Urban public finances in developing countries : a case study of metropolitan Ahahmedabad

DRAFT THE WORLD BANK DEVELOPMENT ECONOMICS DEPARTMENT URBAN AND REGIONAL ECONOMICS DIVISIbN URBAN AND REGIONAL REPORT No. 77-4 URBAN PUBLIC FINANCES IN DEVELOPING COUNTRIES: A CASE STUDY OF METROPOLITAN AHMEDABAD ROY W. BAHL August 1975 These materials are for internal use oily and are circulated to stimulate discussion and critical comment. Views are those of the author and should not be interpreted as reflecting the views of the World Bank. References in publications to Reports should be cleared with the author to protect the tentative charadter of these papers. Foreword This is one of a series of case studies of the,public finances of urban governments in less developed countries. These studies, designed to be as comparable as the local conditions and availability of information permit, are meant to provide a) an understanding of the operation of finance systems in cities of developing countries; b) a format for describing and analyzing these systems; c) a menu of major problems facing these governments; and d) a set of data which may be used to establish comparative norms against which aspects of performance or problem severity of urban governments gener- ally may be evaluated. It should be emphasized that, although each case study contains a section on conclusions and policy implications, the pri- mary goal is not to offer a program of detailed fiscal reform for indivi- dual cities. Rather, it is to provide the beginnings of an analytic and informational base which will pbermit such judgements to be made oh a con- sistent basis and with increasing confidence over time. All but a very small portion of this work was carried out as a World Bank Research Project (RP0270) under the direction of Professor Roy Bahl, of the Metropolitan Studies Program, The Maxwell School, Syracuse University, and Mr. Johannes Linn, of the Urban and Regional Economics Division of the World Bank. The overall project includes six detailed case studies of -bmedabad and Bombay, India; Cartegena and Bogota, Colombia; Jakarta, Indonesia; and Seoul, Korea; plus four more 2imited studies of the public finances of Kingston, Jamaica; and Tunis, Tunisia--as part of RP0270; Lusaka, Zambia; and Manila, the Philippines--as separate, related efforts. In the final phase of this research a set of papers will be prepared on major urban - ii - public finance issues, comparing practices and experiences in these ten case cities, as well as others for which comparable information exists. This case study of The Public Finances of Metropolitan Ahmedabad is an outgrowth of field work carried out in 1972, 1973, and 1974 by Pro- fessor Bahl and myself. Douglas H. Keare Urban and Regional Economics Division Development Economics Department - ii.i - Research Staff Roy Bahl and Johannes Lim= have been responsible for the final editing of all papers issued under--or in association with--this project,. ensuring accuracy and consistency, etc.; however, a large number of other persons have been involved in their production: Detailed Case Studies Princinal Research Arising out of RP0270 Researcher (s) Assistant (s) Ahmedabad Roy Bahl, Douglas Keare Michael Wasylenko, Tamar Katz Bombay Francine Bougeon-Maassen, Michael Wasylenko, Roy Bahl Tamar Katz, Carlos Noble, Mark Gellerson Bogota Johannes Linn, Roy Bahl Michael Wasylenko, Carlos Noble, Tran Than Dang Cartagena Johannes Linn, Roy Bahl Tran Than Dang, Carlos Noble Seoul Roy Bahl, Douglas Keare Hwang Mong Chan, Michael Wasylenko, Tamar Katz Jakarta Johannes Linn, Hartojo Wignjowijoto, Roy Bahl Limited Case Studies Arising out of P0270 Kingston Francine Bougeon-Maassen, Johannes Linn, Roy Bahl Tunis Remy Prud'homme limited Case Studies of Other Origins Lusaka Robert Saunders Manila Roy Bahl, Pamela Brigg Urban-Regional Economics Division Urban Public Finance Project Case Studies PaDer No. Title Author 1 Urban Public Finances in Developing Countries: Roy Bahl A Case Study of Metropolitan Ahmedabad 2 Urban Public Finances in Developing Countries: Johannes Linn A Case Study of Metropolitan Cartagena 3 Urban Public Finances in Developing Countries: Roy Bahl and A Case Study of Metropolitan Seoul Michael Wasylenko 4 Urban Public Finances in Developing Countries: Remy Proud'homme A Case Study of Metropolitan Tunis 5 Urban Public Finances in Developing Countries: Roy Bahl and A Case Study of Metropolitan Manila Pamela Brigg TABLE OF CONTENTS Page I. INTRODUCTION II. GOVERNMENT STRUCTURE AND ADMINISTRATION . . . . . . . . . . . . . .. 2a Local Govertent Structure in India . . . . . . . 2a Federal-State-Local Relations . . . . . . .. . . . . . 3 Political and Administrative Structure of the City . . . . . . 5 Budget Structure and Organization . . . . . . . . . . . . . . 7 III. EXPENDITURE TRENDS . . . . . . . . . . . . . . . . . . . . . . . . 10 General Patterns . . . . . . . . . . . . . . . . . . . . . The Sources of Finance . ....21 IV. TE FUNCTIONAL SOURCES OF EXPENDITURE INCREASE . . . . . . . . 21 Self-Financed Services .. . Water Supply . . . . . . . . . . . . . . . . . Refuse Collection . a . . . . . . . . . 29 Municipal Dair . . . . . . . . . . . . . . . . . . . . 31 Transportation . . . . . . . . . . . . . . . . . 34 General Revenue Financed Services . . . . . . . . . . . . . . 36 V. REVENU STRUCTURE AND TREND . . . . . . . . . . . . . . , . . . . 37 Taxation.. ..... ... .. .... ...... Ll Property Tax . . . . . . . . . . . . . . . . . . . . . . . . . 44 Legal Structure . . . . . . . . . . . . . . . . . . . . . 44 Assessment Procedure . . . . . ,. . . . . . . . . . . . . 45 Administration . . . . . . . . . . . . . . . . . . . . . 2 Revenue Yield . . . . . . . . . . . . . . . . . . . 52 Elasticity . . . . . . .. . . 5 Octroi . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 Legal Structue . . . . . . . . . . . . . . . . . . . . . 60 Administration . . . . . i . . . . . . . . . . . . . . . . 61 Revenue Yield . . . . . . . . . . . . . . . . . . . . . . 61 Other Tax es . . . . . . . . . . . . . . . . , . . . . . . . . 64 -7v- Page VI. iNTERGOVERNNMNTAL ASSISTANCE . . . . . . . . . . . . . . . . .. . 66 General Purpose Grants . . . . . . . . . . . .. . . . . . . . 66 Education Cess . . . . . . . . . . 66 Other . * . -. . . 0 * * * . . . . . 66 Education Assiatance . . ........... ,.. 67 Health Assistane . . . . . . . . . . . . . 71 Police-Justice Assistance . * . * . . . . . . . . . 71 Other Grants . . . . . . . . . . . . . . 72 Equalizing Features of Intergovernmental Flows . . . . . . . . . 72 VII. PROBLEMS AN POLICY RECOMMDATIONS . . . . . . . . . . . . . 76 Budget Structure and Administration . . . . . 76 Expenditure Planning . . . . . . . . . . . . . 78 Tax Reform . . . . . . . . . . . . . . . . . . . 79 Appendix Table 1. A Summary of Revenues and Expenditures in Ahmedabad, 1961-72 (in Rupees) Appendix Table 2. P;er Capita Income in Gujarat State in: Constant and Current Prices for 1960-61, 1970-71 - vii - TAB!ES Page TABLE 1 Urban Local Governments in India . . . . . . . .. . . . .. 3 TABLE 2 Relative Importance of Central, State and Local Governments as measured by per capita total expenditures . . . . . . . 12 TABLE 3 Ahmedabad Local Government Expenditures Patterns: 1965-71 . . . . . . . . . . . . . . . . . . . . . .. . . . 13 TABLE 4 Increase in Ahmedabad Per Capita Expenditures by Function, 1965-1971 . . . . . . . . . ..... .. . . . . . , . . . 16 TABLE . Functions with Increased/Decreased Share of Total Ependitures . . . . . . . . . .. . . . . . 17 TABLE 6 Percentage Distribution of Ahmedabad Local Government Expenditures by Function. . . . . . . . . . . . . . . 20 TABLE 7. Total Expenditure and Percent Distribution of Financing . . . . . . . . . . . .. . 22 TABLE 8 Changes in Total Expenditure and Percent Distribution of Financing . . . . . . . . . . . . . . . . . . . . . . . . . 23 TABLE 9 Water Supply Finances . . . . . . . . . . . . . . . . . . . . 28 TABLE 10 Estimated Difference Between Water Rate Revenue Currently Raised and Estimated Water Rate Revenue Raised Under an Average Cost (Equivalent) Property Tax Pric . . . . . . 30 TABLE 11 Financing Refuse Collection . . . . . . . . . . . . . 32 TABLE 12 Finances and Operation of the Municipal Dairy . . . . . . . . 33 TABLE 13 AMTS Operations and Financing . . . . . . . . . . . . . 35 - viii - Page TABLE 14 Schedule of Fares, 1970-71. . . . . . 36 TABLE 15 Current Expenditure Trends: 1964-1972 . . . . . . . . . . . 38 TABLE 16 Per Capita Expenditures by Functions for Ahmedabad Municipal Corporation 1965-1972 . . . . . . . . . . . . . . 39-40 TABLE 17 Tax Revenue Structure and Trends. . . . . . . . . . . . . . . 42- TABLE 18 Per Capita Tax Revenue, Prices, Per Capita Real Tax Revenue 1964-72 . . . . . . . . . . . . . . . . . . .. . . . . . . 43 TABLE 19 Property Tax Rate Structure . . . . . . . . . . . . . . .6 TABLE 20 Assessment Standards for Owner-occupied Housing . . . . . . . 9 TABLE 21 Growth in Property Tax Base . .. . . . . . . . . . . . TABLE 22 Property Tax Revenue Trends . . . . . . . . . . . . . . . . . 56 TABLE 23 Increases in Net Rateable Value for Property Value Class: 1969-1973 . . * . . . . * . . * * . . . . . . . . . 59 TABLE 24 Octroi Revenue: By Class of Gods . . . . . . . . . . . . . 65 TABLE 25 Rate of the Education Cess: By Rateable Value Class . . . 68 TABLE 26 Grants from the State Government to the Ahmedabad. Local Governments . . . . .. . . . . . . . . . . . . . . . 69 TABLE 27 Revenue from Own Sources, Grants in Aid, Assessed Value of Property, and Population for 52 Municipalities in Gujarat state, 1968-69 . . . . . . . . . . . . . . . . . . . . 75 - ix - FIGURES Page Figure 1 Organizational Structure of AMC Government . . . . . . . . . 8 4 TEE PUBLIC FITANCES OF METROPOLITAN AEMABAD I. Introduction Metropolitan Ahmedabad, though the largest city in Gujarat State with a population of approximately 1.6 million, accounts for only six per- cent of the total State population. The Ahmedabad metropolitan area covers 124 square kilometers and has an overall population density of about 13,000 per square kilometer but within the walled area at .the core of the city over 250,000 people live in a two square mile area. By comparison with the rest of Gujarat State, and with other large urban areas in less devel- oped countries, the population of Ahmedabad is not growing rapidly. The economic base of the area is dominated by manufacturing--over half of urban employment is in the manufacturing sector as opposed to 27 percent in trade and services. This manufacturing activity is dominated by the textile mills, which have long been the major local employer, and as might be expected, the local governments in Ahmedabad are heavily dependent on the textile industry for revenue generation. Hence to a large degree, the fortunes of Ahmedabad's local public sector are tied to the external markets of the local textile mills. The major local government in the Ahmedabad metropolitan area is the Ahmedabad unicipal Corporation (AMC). However, the AMC is over- lapped by the Ahmedabad Metropolitan Transit Corporation (AMTS) and the School Board, neither of which are fiscally or administratively inde- pendent of the AMC. 2. Ahmedabad, like alll big cities in the world has pressing expen- diture needs and inadequate government resotfrces, i.e., is subject to a problem of public sector poverty. In Ahmedabad, housing and utilities are particularly severe problems with one-third of the urban area popu- lation living in I'slum" conditions with inadequate housing structures, overcrowding, few public service amenities, and poor sanitary conditions. Even so, one would not characterize Ahmedabad's local government fiscal position as "critical", in the sense of pending bankrupicy or the ur- gency of immediate and drastic fiscal reform. On the other hand, there is a slowing down of revenue growth which could re8ult in a serious future revenue-expenditure mismatch. Moreover, while the current sit- uation is not one of crisis, there clearly are fiscal problems and areas where reform could bring substantial benefits to the local population. Four major areas where policy action is needed might be mentioned brief- ly here, though each is explored in detail in the text and recommendations sections below. First, there is need to increase the elasticity of the property tax through structural reform, not only to meet current and new expenditure needs, but also to begin to clear up the backlog of infra- structure and service needs which plagues particularly the city4s low income population. Second, there are both horizontal and vertical in- equities in the property tax which need reform. Third, certain of the government "enterprise" type activities are operating with deficits and require adjustment. Fourth, there is a paucity of systematic fiscal planning. The formulation of remedial policy to deal with these prob- lems and the evaluation of other facets of the operation of the ur- ban public finance system require detailed study of the overall func- tioning of the system. Such detailed study is the purpose of this research.. I. Government Structure and Administration Local Government Structure in Indial/ Local government matters in India are primarily the resPon- sibility of the 15 State governments. Apart from a general advisory role, the Central Government exercises no direct control over the local governments. All local governments (with the exception of military cantonments: are the creations of State government legis- lation. Local governments in India may be divided into Panchayats, which operate in rural areas, and =unicipal bodies, which operate in urban areas. There are about 2,000 urban local bodies in India, as opposed to 200,000 rural local governments (see Table 1). A brief summary of Indian local government structure is to be found in A. R. Marshall, Local Government Finance, The Hague: International Union of Local Authorities, 1969. (Report prepared for the 1969 IULA Congress in Vienna.) A more complete summary is contained in the ITUrban Public Finances in Developing Countries: A Case Study of Bombay" (Urban and Regional Economics Division, IERD), forthcoming. 3. Table lä URBAN 10CÄL GOVERENTS IN INDIA Municipal corrations 29 Municipal councils 1,483 Cantonuent boards 62 otified area committees 1 `1 To-m area coittees 3,27 Source: A.H. 'arshail, co Governnt Finance, The Hague:. Itr Uion of' Local AthoritieE, 1969. (Rerort prepared `Eor- the 1969 rLA Congress i.n Vienna.) Of the 2ive types of urban local authority in India, only two- imicipal cor:;orations and muniicipl council~-re fUJly representative local government unnits. Cantonr.ents are permanent rilitary stations with, partially norinated boards headed by the officers commanding the stations. The notified area corittees are fully nominated and are meant for growving tows wrhich are not yet ready for fu.l mni cipal government. The tow- area corraittees are set up foor small twms with ILnited functions. The rruici;al govemnment in Ahmedabad. is a corportion, created under the Bombay Provincial and -unicial Carroration Act in 1950. There is a military cantonnent ;i+hin. Ahmedabad, but as is the case in all of India, it iv autonomiiously governed and the AMC has no jurisdiction. Federal-State-Lccai RClations Thae attern of intergovernmental relations (fiscal and regulatOry) in Gujar:at, as in all Indian Statcz, is one of close Central-State and State- Local ties. The Central Government exerts no direct regulatory controls over local government bodies and direct Central-local government loan or grant programs are limited to slum clearance efforts and to some categorical assistance in the health field. Indirectly, however, the central government effect on local government activity may be substantial. These indirect effects are of two types: (a) the central government makes substantial (direct) expenditures in the urban area, primarily on railroads, roads, and higher education, (b) there are certain mandates associated with Federal to State grants in India, which concern the "pass-through" of Federal assistance to local units and which may affect the amount of State government resources allocated to the urban areas. Another indirect effect relates to debt. The Central Government sets a limit on the amount which all governments operating within a State (including the State government) may borrow. The State, in turn, allocates borrowing authoriza- tions to local governments within the constraints of this limit. The intrastate allocations are strictly a state government policy decision. State-Local relations are more direct and important in terms of AMC operations. The Gujarat State government exerts a set of direct regulatory controls over the AMC, gives grants and loans to the AMC, and makes substantial direct expenditures within the urban area. The most important of the direct controls over the AM.C by Gujarat State are approval of the AMC budget, approval for tax rate increases beyond the prescribed statutory limits, approval of borrowing from the public, and appointment of the chief financial officer of the City Government (the Municipal Commissioner). Political and Administrative Structure of The City Though the principal elective office in the AMC is that of mayor, the mayor has little executive, or administrative power or authority. The city council, or general board, is composed of 91 elected members, of which 81 are elected from 33 election districts of various sizes, and the remaining 10 are elected at large to represent scheduled tribes and castes The Mayvr is elected by the 91 councillors and along with the councillors, serves a term of five years. The Bombay Provincial -Lzd Municipal Corporation Act (1950), under which the AMC was created, prescribes five governing bodies within the Corporation, and the Bombay Education Act of 1923 (revised 19L7) has added a sixth and seventh. The three which are mandatory are the general board of 91 members, a 12-member standing committee which is elected by the general board, and a municipal commissioner who is appointed by the State goverment. The Municipal Commissioner is advised by the standing committee and is assisted by three deputy commissioners. The Bombay Municipal Corporation Act also provides for, but does not require, a nine member transport committee and a transport manager. Both of these discre- tionary units exist within the AKC. Finally, a separate piece of legislation-- the Bombay Primary Education Act--provides for a 14-member school board and an administrative officer for primary education. The transportatlion coand the school board enjoy substautial (but not comiplete) autonorny fron the municipal comissioner. This autonomy results partly from the prescribed functions of the Comm4ittees and the Comissioner, and partly frorm the f act that education and transpurt (bus) finances are not wholly within the `C budget. Apart from the cormmlittees and positions which are provided by statute, there are eig,t special comhittees which have been created by the General Board of the AmC: (a) Healtn (b) Recreation and Culture (c) Legal (d) Public Works (e) Elousing and improvement (f). mi lk (g) Home Planning (h) -Refuse and Sewerage Councilmen are eledted by the Board to these com.ittees and elect a committee chairman from the membershio. The day-to-day administration of the MIC is tne responsibility of the Hunicipal Commis.sioner. He is assisted by three deputy commissioners-- for special matters, general matters, and revenue and legal mätters-and by a City Engineer, Chief Accountant, City Health Officer, and a Town Develo-tment Offiier. The Deputy Commissioner for Special Matters has responsibility for assonbling. the budget, and also for plarrdng, housing, and estate management. The Deputy Commissioner for General Affairs has responsibility for Public Tirks, Health, and the general control of city goveriment. The Deputy Co.issioner for Revenue anda:. r is concerned with tax assessment and collection, and w.rith cllective 'arganing. An interestin point i this di-ision of resonsibility is the separation of revenue analysis and estim.ation from the Deputy Co-rissioner for Special Matters who has resronsibility or the budget. The overall administration of the AMC is described in the organization chart presented as Figure 1. Budget Struct-ure Pnd r ni -ation The budget is prepared initially by the 'unicipal Commissioner and is presented first to the Standing Committee and then to the General Board. The final budgc-et is approved by the General Board, which can change budget estimates as long as a 100,000 Rs cash balance is kept. The badget of the AC is com-Drised of a revenue budet and a carDital works budget, On the ex-c-nditure side, the revenue budget includes all ordinary etUenditures, all debt service expenditur es, a ca.pital expenditure component, transfers to the school board and bus company, and all expenditu-res on the city milk scheme. On the income side the revenue budget includes all current revenues raised from own sources, grants and contributions, reverues from the ni.ilk scheme, and proceeds from borrow,ing, either for the. miIk scherie or for the small arount of capital expenditures included vrithin the revenue budget., The latter, general revenue capital income, is the balancing item in the ANC revenue budget. Most capital expenditures made by the AMC and all those firnanced from locs are shown in a separate capital budget. The revenue ite-,s in this capital budget are payments from reserve accounts, payments from the revenue budget, and rroceeds fron, borroving. Figuro 1: ORGIANZLTIONAL STMJCTUPJE OF AMG GOVERiIMENT Standing Committee l M1ayor j Gneral Board Adminis trativo Administrative Inicipal Officer fLr School Board Officor for Transport .jnicipal . nicipal Chicf School Board Tranaport Committee SocrCty Comissionar Auditor Deputy Deputy Deputy Comissionor , Cormissioner Co:rissioner Special Araira General Affaira Rovaems and 11tý'Lav Public rks1 Hoalth Rocreation and Town tcInal Housing and Refuse and Milk Committeo Conmittee Cultural Connittoo Planning Devlopment Sowerage Towm Dovalopment Chief City City officar Accountant. igincer Health Officer !1 9. Because of the budget format and presentation, functional expenditure analysis using budget and financial report data is difficult. Expenditures in the revenue budget are presented by object (e.g., wages and salaries, repairs, etc.) and expenditures in the capital works budget are often presented by program (e.g. the Dharia project). However, the AMC does prepare, yearly, a special statement of "ordinary" revenue and expenditure which enables a functional rather than object breakdown. No breakdown exists for "extraordinary" revenue expenditures (e.g., revenue capital expenditures), or for the capital expenditure budget. The separation of current from capital expenditures is easier because there are no recurrent expenditures in the capital works budget and the revenue budget inclu'des capital expenditures under the heads "revenue capital expenditures", and "capital subsidies to the AMTS". The estimation of a consolidated local government budget for the Ahmedabad Urban area may be made by summing the budgets of the AMC, the School Board, and the Ahmedabad Municipal Transport Service (AMTS), after appropriately netting out transfers between the AMC and the other two local bodies. It should be underlined that there is an absence of any long term fiscal planning for any of the three local governments in the urban area. Beyond the one-year budget, there are neither planned expenditcre levels nor revenue projections, and there is no office within the City government which has fiscal planning responsibility. 10. III. Expenditure Trends The purpose in this section is to describe,and, if possible, explain the pattern of increase in Ahmedabad local government expendi- tures.1/ Three considerations are relevant: the general pattern and trend of expenditure increase, the sources of financings this increase, and the functional components of this expenditure increase. Throughout, there is special emphasis on the extent to which any expenditure activities generate revenues which account for some part of costs (are part:ially or wholly self-financing) and therefore on the general revenue dependence of each expenditure category. General Patterns As noted above, public expenditure activity within the Ahmedabad metropolitan area is assignable to one of the three units of local government (AM:C, AMTS, or School Board), to the Gujarat State government, or to the Central Government. Neither the Central nor the State government keeps records on direct expenditures within the Ahmedabad urban area, therefore it is not possible to quantify their relative expenditure importance and to compare this with local government exenditures.-A crud.e estimate of the relative importance of the three levels of government in the urban area might 1/ Throughout this paper, the term "Ahmedabad local government expenditures" will be taken to include the financial operations of the municipality, the school board and the ANTS, as opposed to the term "AMC expenditures ffwach will be used to refer solely io the Municipal Croration. 2/ Even with a larger expenditure of effort it would not be possible to assemble these data. Expenditure.records are kept by purpose of expenditure, oy function, or by object, not by location of the expend-itUre. be made by comparing their overall levels of per capita expenditures, i.e ., by comparing per capita expenditures by the Government of India (less transfers to State and Local units), per capita expenditures by the Gujarat State Government (less transfers to local governments), and per capita expenditures of local governments in Ahmedabad (less intra- local transfers).1/ Such a comparison is presented in Table 2 and shows that Ahmedabad local governments spent about as much per capita as the other trw levels of governments combined in 1965, _The importance of State and- National government expenditures, however, has increased between 1965 and.' 1971.. Per capita expenditures of the Gujarat State Government and the Indian Union Governet grew at average annual rates of 12.7 and 11.5 percent respectively, while per capita expenditures of Ahmedabad local governments grew at 'the much lower average annual rate of 6.1 percent. The time series data in Tacla 3 shows that the growth of local government expenditures within the Ahmedabad urban area has been steady. Data limitations are such that a systematic study of the dete=minants of this 1/ Strictly, the assmption that this would measure relative imortance requires assmning, for exmple, that average nationwide per capita spending by the Indian Government (State of Gujarat) is equal to average per capita spending by the Indian Government (State of Gujarat) in the Ahmedabad area. If, for example, the Indian Government spends, in Ahmedabad, an amount less than the national per capita average, then the relative expenditure importance of local government is understated in this comparison. 12. Table 2: RELATIVE TIPORTANCE OF CENTRAL, STATE AND IDCAL GOVERNMENTS AS MEASURED BY PER CAPITA TOTAL EXPMDITURE (in Rupees) Per Capita Amount Average Annual Average Annual Percent Increase Percent In- 1965 1971 In Per Capita drease in Expenditures Porulation 3/ Ahmedabad local 100.9 143.2 6.1 3.3 Governments 1/ Gujarat State 52.6 108.1 12.7 2.7 Government 2/ Indian Union 48.1 92.4 11.5 2.2 Government 2/ TOTAL FOR 201.6 343.7 9.3 3.3 AHMEDABAD AREA 1/ Including net expenditures of the School Board and AMTS 2/ Excluding transfers to lower level governments 3/1 1961-1971 Sources: Quarterly Bulletin of Economics and Statistics, BLreau of Economics and Statistics, Government of Gujarat, Ahmedabad 1964-65; Socio-Economic Review, Gujarat State 1972-73. Bureau of Economics and Statistics, Government of Gujarat, Ahmedabad; Documents of IBRD and IDA, Economic Situation and Prospects of India, May 8, 1973, June 20, 1966. Tabl e 3: AHMEDABAD LOCAL GOVERNMENT EXPENDITURE PATTERN: 1965-71 (in Rupees) TOtal Yeal Per Capita utal E:cxndi Ures Ratio of Current to 'Total 1/ -n:nditures 2/ (1965 Prices) Total Epndliture Expenditure Poulation Prices 1,321,61,2it6 1,321,61,216 .761 100.9 1,309,387 100.0 i166 1,hQ6,99,701 1,379,72,074 .803 110.7 1,352,597 108.5 1j,7 1,r00,90I 1,158,99,954 .880 107.4 1,397,233 129.5 1-63 1 797,61,265 1,105,68,291 .836 124.5 1,443,31t2 127.9 6 1 93, 3 ,118 1,506,36,384 .859 131.0 1,490.972 129.5 1976 2037 ,3,2C3 1,501,3h,268 .875 132.3 1,540,174 135.7 1971 2,27),2,272 1,648,19,472 .880 143.2 1,588,379 138.0 3' Average- . aL 9 3.7 6.0 3.3 5 1/ Estimated using growth rate between 1961 and 1971. Total Expenditure includes Expenditure on Milk Scheme, Transportation and Education. Throughout this study, average annual growth refers to the compound rate of growth between the first and last year shown in the data. Sources: Ahmedabad Municipal Corporationt Budget Estimates, 1965-71 and "Press Notes". Socio-Economic Review of Gujarat State 1972-73, Consumer Price Index Numbers for Industrial Workers in Ahmedabad (converted to 1965 base), p. 71. growth is not possible. However, the trend in per capita expenditures might be viewed as generally effected by increases in the price level and increases in population.-/ Had per capita total expenditure remained con- stant in real terms at Ra 100.9 between 1965 and 1971, total money expendi- tures would have risen to R 221,169,069 or by 93.L percent of the_at Increase,R ois oly- 6-.6 percent of the increase is due to increases in real per capita expenditures. Using this approach to explain the change in total ex- penditures,_29 . is the percentage expenditure increase which may be assigned to rising population, 16.5 to increased per capita spending at the original levels of population and prices, 52.7 to inflation, and 1.3 percent to the interaction of all three factors. Still to be dealt with is the question of whether the increase in per capita spending which did occur is primarily a reflection of an u=grading of services or if it primarily reflects incr--s- ing:wage rates. The six percent annual increase in per capita expenditures of Abmedabad local governments over the past decade is primarily due to heay increases in current expenditures. Capital spending bas fallen from about 2h percent of the consolidated budget in 1965 to 12 percent in I/ The change in expenditure might be viewed as a total differential of total expenditures (E) with respect to population, per capita expenditure and inflation, which are denoted P, E, I, respectively; i.e. dE - (P + dP) (I + dl) ( + dR ) More specifically P, I, and Z, represent population, the index of 'prices and real per capita expenditure rtspectively in the initial year and dP, dlf, and dE is the change in those variables between the initial and the final year. Tn this analysis the initial year is 1965, the final year is 1971. The last term in the differential expression E(PEpl) is the total exbenditure in t1e initial year, which is subtracted from the final yeaLr expenditure io obtain the change in expenditure. The price index is set at 1.00 in the initial yv7ear. Al-he radables_a.e_kpo.wn .exce t the real change in per capita expenditure, for which we solve the differential. The procedure is described more fully in Appendix A. 1971 (see Table 3). That this decline has been relatively consistent over the past decade may be seen from the more detailed trends presented in Table 4. One explanation for this pattern is a desire on the part of local government officials to increase the level of spending for the social service functions, i.e., these functions are less capital intensive and.increases in spending for them would be reflected in heavy current expenditure increases. In fact, there were above average increases in per capita spending for education, welfare, fire protection, transportation, and the Milk Scheme (see Table 5). The latter is financed heavily from user charges, and the expenditure increase may therefore be explained in part by the ability of the AMC to expand milk sales. Except for transportation, the other functions are financed through general revenues, and in the case of education, about one-third through grants-in-aid. The more heavily capital-intensive functions--water supply, sewerage and drainage, and roads and bridges--all grew at a slower rate than total local government expenditures (Table 5). As a result of these differential growth rates, the functional distribution of expenditures has been accentuated toward expenditures on the Social Service Tunctions.-/ As may be seen from Tables 5 and 6, social service expenditures have increased from 55.9 to 67.2 percent of total spending. Though the magnitude of this increase is due in-large part to the activity of the Municipal dairy, social services would have increased from 53.2 to 59.2 percent of total percent of total spending even without the consideration of the Milk Scheme. In exploring The very broad definition of social services used here includes Health, Education, Welfare, Fire, Refuse Collection, Housing, Transportation, the Milk Scheme, Markets and Recreation. Table h: INCREASE IN AHMEDABAD LOCAL GOVERNMENT PER CAPITA EXPENDITURES BY FUNCTION: 1965-1971 (iin rupees)- Annual .1Annual Annual Percent Percent Percent FUnctien 1965 1971 Increase Fumction 1965 1971 Tncrease Function 1965 1971 Incrase ulter (total) 6.2 8.1 1.5 Fire (total) 0.7 1.6 11.8 Land Development Current; 2.5 3.6 6.3 Current 0.5 0.9 10.3 Tot: 1.1 CapIti 3.7. 4.5 3.3 Capital o.2 0.7 23.0 Current Capital 1.1 S Seierage & Refuse collection Draii.,ge Total 1.0 4.8 3.1 Yilk Schene Total i 5.0 6.5· 4.5 Current h.0 4.8 3.1 1Ib tal 5.8 27.9 30.0 CurrenL 2.5 3.1 3.7 * Capital - - - Current 5.6 27.5 31.0 Cäpi Ual 2.5 3.4. 5.3 Capital 0.2 0.1 12.3 General Adminis- Rbads & bridges tration Marke t (total) 0.3 0.2 -7.0 To t-al 7.4 6.3 -2.7 To tal 5.7 8.2 6.2 Current 0.15 0.2 ,.9 Currenl 3.5 3.9 1.8 .Current 5.3 8.1 7.3 Capitai 0.15 0.01 -55.0 Capital 3.9 2.1 -8.h Capital 0.14 0.1 -26.0 Recreation H:alth (total) 8.9 11.3 4.1 Housing (total) 5.6 1.1 31.0 Total 1.3 1.7 1.0 Current 8.h 11.1 h.8 Current 0.2 . Current 1.3 1.h 1.2 Capital 0.5 0.2 -16.5 Capital 5.h 1.1 . 31.0 Capital 0.5 -0.3 -9.0 Education (total)11.8 23.6 8.1 Transportation Other (total) 8.2 2.6 -21.0 Curren L 12.8 22.8 10.1 Total 13.7 '22.3 8.11 (rrent 3.6 0.5 -39.0 Capital 2.0 0.8 -16.5 Current 13.7 22.3 . 8.4 Capital h.6 .2.0 -1Ih.0 Capital • - - - mifare (total) 0.3 0.6 12-3 TOTAL 100.3 113.2 6.1 Curren t 0.3 0.6 12.3 Debt service Current 76.2 126.1 8.8 Capital - - - Total . 11.8 15.3 h-5 Capital 2h.1 17.1 -5.8 Current 11.8 15.3 4.5 Capital - - 17.. Table 5: FUNCTIONS WITH INCRASED/DECREASD SHARE OF TOTAL EIPENDITURES Increase -n Tecrase in Increasing Percentage Decreasing Percentage Functin: share Rinctions share E,ducaitionx 1.7 våate r 0.6 Welfare 0.1 Sewerage & Drainaga 1.9 Fire 0.4 Roads & Bridges 3.0 General Admrinitration 0.1 Refuse Collection 0.7 Transportation 1.9 Housing 4.8 Land redevelopment 0.8 Debt Service 1.1 Milk Scheme 13.7 -Health 1.0 18. the general trend of Ahmedabad local government expenditures, another question which might be raised here is how have expenditures, and revenues from own sources, increased with p6pulation increase (see Appendix Table- 1) i.e., have the Local goverments become relatively more or less self-sufficient. The appropriate relationship can be expressed as follows: R = a1 p (1) E 2 P (2) where R = revenue from our sources in the AMC for years 1961-1972 P - population bf the AMC for years 1961-1972 Z= Current expenditure in the AMC for years 1961-1972 al 2 constants are le elasticities of revenue from own sources and current expenditure with respect to population, res- pectively. From equations (1) and (2), the following relationship can be derived: R III P (3) Sa2, Equations (1) and (2) are derived here by taking the logarithms of R, E, and P and estimating with ordinar east squares- Estimationof (3) is obtained by-substitution. The results show: log(R) -10.1131 + 2.92 log(P) R .97 (L) (16.9) log(E) -9.9069 + 2.89 log(P) 2 .97 (17.8.) and, by substitution, R/E = 1.02 PO.03 The t-values, shown in parenthesis, indicate that the elasticity parameters . . 19. for both revenue from own sources and current expenditure with respect to population are significantly different from zero at the 0.001 level of significance. The fact that both elasticity coefficients exceed unity means that both expenditures and. revenues from own sources grow proportionately faster than population. Mre preciselF, one percant increan a iz -opillation induces about a three percent incre&e in both revenia from own sources and current expenditure. Accordingly, the fiscal "self-sufficiency of Ahmedabad local govern- menta has remained appro=imately constant over this period. 1/ Here, and throughout this paper, all time series estimates must be inter- preted with great caution. The small number of degrees of freedom eff- ectively prohibits our doing any corrections for autocorrelation or making adjustments for multicollinearity; Rather, we have simply pre- sented the results as the best one can do given these data limitations. 20. Table 6: PERCENTAGE DISTRIBUION OF AEABAD LCCAL GOVERNMENT EXENDITURE: BY FUNCTION 2965 1971 F4nction urrent To tal Current Tta Water 3.2 6.2 2.9 5.6 Se:wrage &drainage 3.2 5.0 2.5 3.1 Roads & bridges 4.6 7.4 3.1 4. Health 11.0 8.9 8.8 7.9 Education 16.9 1h.8 18.1 16.5 Welfare 0.4 0.3 0.5 0.h Fire 0.6 0.7 0.7 1.1 Refuse collection 6.h h.0 3.8 3.3 General adminis- 7.0 - 5.7 5.8 tration Housing 0.2 5.6 0.8 Transportation 18.0 13.7 . 17.7 15.6 Dabt service 15.5 11.8 12.2 10.7 Land redevelopment 0.8 Miik Scheme 7.4 5.8 21.8 19.5 Market 0.2 0.3 0.1 -0.1 Recreation 1.7 1.8 1.2 1.2 Other. 4.8 8.2 1<47 1.8 Total . 100. 0 100.0 . 100.0 100.0 1/ Totals may not add-to 100 percent due to rounding. 21. The Sources of Finance A key to understanding and explaining the.trends in local government expenditures is measurement of the extent to which the various sources of revenue -- taxes, user charges, non-tax revenues, external aids, borrowing -- have contributed to the financing of expenditure increase. The "average" structure of finance for Ahmedabad local government expenditures is characterized by heavy reliance on general tax revenues, and on user charges from the Milk Scheme and the Bus Company (see Table 7). There is but minor dependence on Central and State government finances. When the financing of 1965-1971 changes in Abmedabad local government expenditures is examined, it may be seen that this pattern is generally reinforced, i.e., nearly 40 percent of the expenditure increase was financed through general taxation and most of the remainder through user charges under the self- financed services (see Table 8). IV. The Functional Sources of Expenditure/Increase Another approach to explaining,Ahmedabad local government expenditure trends is to focus on expenditure increases on a function basis, with srecial concern for the trends and causes of such increase, and for identifying any part of'-expenditures which are covered from resources other than general purpose revenues raised from own sources. Accordingly, specific attention here is accorded those functions which recapture a portion of- expenditures through specific charges or benefit levies, and expenditure functions are viewed according to their financing arrangements, i.e., general revenue financed, or "self -financedn in cases where the function is supported either wholly or in part by a user charge or earmarked tax. Table 7: TOTAL EXPENDITURE AN4D PERCENT DISTIBUTION OF FINANCING- (in Rupees) 1965-71 1971 1970 1969 1968 1967 6 1965 Total Expenditure 12,382,70246 '2,274,92,272 . 2,037,32,203 1,953,33,118 1,797,61,265 i,5o0,9o,4i 1,486,99,701 1,321,61,246 Percent Distribution, of.Financing Tax Revenues 40.1 38.6 ,43.5 '3.7 32.2 39.5 33.9 38.7 S Non-Tax Revenuee 6.5 5.9 6.2 5.9 5.8 6.8 6.8 7.1 Grants 3.5 4.2 3.5 2.9 2.9' 2.8 3.0 '4.1 Louns 13.5 11.0 11.9 L2.3 16.0 11.8 15.7 .14.6 Self-Financed 36.4 .41.8 39.4 36.6 34.6 34.6 29.0 24.4 Table at CHANGES IN 'OMTAL EXPEDI'UE AND PEPE' DISTRIBUTION OF FIl01C ING (in Rupecs) 1971-65 1971-70 1970-69 1969-68 1968-67 1967-66 196665 Change In Total 953,31,026 237,60,064- 83,990,850 155,71853 96,70,824 13,90,740 175,33,455 ExpMnditure Percent Finnceöd Tax fronnes 38.4 -4.2 39.0 174.0 -3.2, 638.6 -2. Non--Tax 1kvenues 4.1 3.2 11,9 . 7.0 1.1 -0.4j. Gra-xn ts 4.3 9.7 - 17.4 3.8 3.3 . -18.6 - Loan. 5.9 3.1 4.0 -31.1 37.5 -405.5 23.7 Sel-Financed 65.9 62.7 103.9 59.9 34.6 633.7 64.0 1/ A percont in any column equals the Increase 1t tha Revenue Category between o.e two year4 divided by the increase in expenditurec. The percentagc distribution' n'eed notd total 100. As noted above, a functional breakdown of Ahmedabad local govern- ment expenditures is complicated because the budget is not structured along functional lines and because there are no detailed expenditure data for in- dividmal subfunctions. This is particulary true for capital expenditures, where an amount for a maltifunction project is often shown in the aggregate with little or no possibility for a breakdown. Ihere such problems are encountered, simplifying assmptions had to b:e made, These are noted below in the appropriate sections. Self-Financed Services The functions of the Ahmedabad Local governments may be roughly divided into those which are self-financed (levy a user charge), and those which are general revenue financed. The AMTS and Municipal Dairy clearly belong in the former category. The water supply and refuse collection fund- tions, though in practice general revenue financed, do have "designated,? revenue sources. For purposes explained below, the Water and Refuse func- tions are treated initially as though they were "self-financed." All other functions are assumed to be general revenue financed. Water Supply. Water supply in Ahmedabad is an AMC responsibility. Water is (officially) provided to about 72 percent of total households in the urban area, with in-house supply for about 66 percent. Average consumption (per resident) has grown from 13,.9 gallons per day in 1961 to 15.7 in 1971. Water is supplied from river (38 percent) and tubewell sources, and total system leakage is estimated at only about 5 percent of water production. There is very little non-residential use of AMC water since most industrial users have their own supplies. The growth in water supply expenditures (shown in Table 9) is due primarily to heavy capital spending and the ensuing'debt servicing obliga- tions which have resulted from expansions of the city's drinking water supply. 25. The financing of water supply expenditures is from general revenues and from borrowing. There is a 2.5 percent property tax rate for residential water supply; however, this designation of the revenues is apparently more a justification for imposing the surrate than an intended means of self- financing. That is, total water expenditures are not limited to this amount, and the user charge is not based on consumption. In fact, the designated water tax has represented a declining share of total expenditures for water siupply forcing an increased reliance on general fund revenues other than the designated water tax, and on borrowing. Since the water tax rate has remained constant for a number of years, the slow growth in designated current revenues for water supply financing is due to the more general problem of a slow_growth in the valuation of prqperty. Because further extension of the system is planned for the next few years, the shortfall between water rate receipts and current expenditures for water supply will increase, and therefore increasingly more general fund revenues will be required. If the increasing drain on AMC revenues for water supply financing is a result of a low property tax "price" for water, then an upward price adjustment might be proposed.: In this respect, it is possible to estimate that water-property tax'rate at which all currnt water supply expenditures are covered. It is important to note that an increase in the designated property tax rate for water financing in no sense improves efficiency, since there is no relationship between the water charge and consumption. The water-rate price equivalent may be estimated if only assessed property value and the average cost of providing water are known. The latter may be approximated with available statistics. Operating and capital costs are Dresented in Table 9 and though actual water consumption/production data are 26. not available for all years, they are roughlT estimated by- interpolating 1961 and 1971 production. Estimating a linear homogeneous total cost function using ordinary least squares, we obtain an equation with a significant negative fixed cost terMI_--a result which makes little economic sense in this case. Because of this result, which is assumed here to be an outgrowth of data inad- equacy, we use only a point estimate of average cost, i.e., cost per gallon in the most recent year for which data are available. In teims of revenue adequay, total property tar revenue. designated.-- for water supply financing would have to be exactly equal to total current expen- ditures. Such an amount may be easily determined, ex oost for the 1970-1972 period. To raise revenues to a level which covers water costs, the average property tax rate designated for water supply financing would have to be increased from its present level of 2.5 percent to 3.5 percent if net rate- able value were to remain constant (see Table 10)., For example, 1/ = -48.64 + 0.0116Q. (R = .96) 2/ Originally we forced a parabola to the data, and estimating with data for 12 years using ordinary least squares derived the following relationship: E - 9.60 - o.oo4 + 0.000001Q2 2 .98 (-.470)-' (1.84) where Q is water production volume E is total current expenditures on water supply, net of depreciation The figures Ln parentheses are the t- values. These values of t with nine degrees of freedom indicate that neither of the parameters is significantly different from zero at the 0.05 level of significance. Clearly there are serious estimation problems because of the time trend build into the independent variable. Since only a rough approx- imation is sought here, thii estimated cost function may provide appropriate information. 27. in 1972, the total expenditure for water consumption was Rs 6.2 million, but the designated property tax surcharge for water yielded only Rs 4.9 million (see Table 10) a shortfall of 27.5 percent. A property tax rate of 3.5 percent, instead of the existing 2.5 percent would eliminate this shortfall. With appropriate overall property- tax reform, the required rate increase may be considerably less. Such reform relates to an equity problem which arises under the existing property tax system. .Byconvention, the assess- ment of owner-occupied residential property is preferential and is esti- mated to be about one-fifth of the assessed value of rented residential - property (for detail see the Property tax section below). Hence owner- occupiers, who are the higher income in the Abmedabad area, pay on average four-fifths less for their water. Moreover, they probably receive better service, because of in-house taps and fewer _eriods of intermittent supply. If the assessment inequity were removed by equalizing the rateable value per Rupee of market value of owner-occupied residential property, the required average property tax rate for water supply financing would be lower. A rough estimate of this rate is 2.0 percent which would in fact lower the current water sur=ate from 2.5 percent.1/ 1/ This estimate is derived by assuming (a) that the percent of residential net rateable- value attributable to owner- occupied housing is equal to the actual percent of owner occupied housing (23 percent) and (b) that this portion of residential rateable value may be increased by four-fifths. Specifically if 81.9 percent of the net rateable value is residential (c.f. Table- 20) and 23 percent of the residential property is owner occupiedthen (.23)(.813) or 18.7 percent of the total net rateable value of residential property_can be assessed at five imes its present assessment rate. Thus letting the present net rateable valae be represented by 1, the total net rateable value can be raised by (5)(.2*3)(.8i3)(X)* +(.7)_(.813)( + (18.7)(X) = 1.748(x) or 174.8 percent of its current value. Teable 9 WA'NRi SfJPPLY FINECES tliRupe"s) 1971-72 1970-.71 1969-70~ 1%S--69 19676 i1/-62 196 -66 196-6 19634.61 1962-6) _ 961- 196')-61 Tetal, Water Charges U,93,191 L7,05,703 51,9Y,376 53,22,c65 23,0014 23,97,736 30,83,562 30,82,757 27,,268 25,36,505 23,77,6is 21,17,33 Total CurrGnI.1,1540 2,1,0 1,21~ ExTot n di turc3 62,3;,U 57,76,197 b9,76,270 U1,28,875 3,50,506% 32,29,693 36,28,656 32,21 549 26,93,139 24,61,14o 22,24,300 Cur-rent Surpiun/ -~1,5,147h -9,90,1h97 .2,18,023 ~,2,3,190 -23,149,163 -3,31,962 -5,U4,6914 -1,30,792 91,129 65,366 1653,58 1,3't,9103 Dceficit Capital Expanditureg! 35,94,00 70,53,000 74,10,00" 28,36,030 39,07,) 23,65,000 36,23,000 ho,92,0O3 Overall Surplu/ 49,40,003 -30,43,497 -71,91,02 4-i9,42,810 -59,:;6,163 -26,96,962 -41,67,694 LO,26,634 Financ-!ngI Borrowing | 15,94,000 70,53,000 74,10,000 28,3,000 39,07KX 23,65,000 36,23,0 \ 1,38,792 .40,92,000 Genoral Revenue3 13,45,474 -9,9o,497 0,49,163 3,31,962 5,14,694 1 Borrowing is OstiMatod au total capital exper.ditures. - 0-i 29. Refuse Collection. Expenditures on refuse collection, or con- servancy, incJude garbage and nightsoil collection and'road cleaning and watering. These expenditures are not supported by a user charge, however, there is a conservancy tax (a property tax surrate) levied at an average rate of 3 percent of net rateable value. As may be seen from Table.11, these receipts fall short of covering costs and because of the low growth i.n property valuation there is reason to expect a further growth in this disparity. To meet this shortfall, the designated property tax rate might be raised. As may be seen from Table 11, the property tax rate would have to be raised by 1.2 percent over its current level to cover full costs. With an equalization in the assessaent ratio, as described above, the required increase would be only 0.5 percent. Another possibility for financing rising conservancy charges would be to make use of a system of user charges. A system could be designed so as to cover costs and reflect differential abilities to pay. 30. Table 10: ESTIMATED DIFFERENCE BETWEEN WATER RATE REVENUE,CURRENTLY RAISED AND ESTIMATED WATER RATE REVEUE RAISED UNDER AN AVERAGE COST (EQIVALENT) PRPEITXTd CARGE (in Rupees) Revenues Raised: if Property tax rate 1972 1971 1970 were equivalgnt to average cost 6,238,663 .5,776,157 L,976,278 With current rate 4,893,191 4,785,700 5,194,376 Difference Amunt 1,345,474 990,457 2.8,098 Percent 27.5 20.7 -.2 Necessary Effective Property Tax Rate to Cover Costs With Existing Rateable Value 3. 3.4 2.8 With Equalized Rateable Value Ratios for Owner-Occupiers and Renters 2.0 2.0 1.7 Municipal Dairy. A dairy enterprise is operated by the AMC, on a partially self-sufficient basis. The dairy has a. milk production capacity of over one lac liters per day, and. is presently operating at that capacity. Ice- cream.is also produced, as are some quantities of cheese and butter. TherS! are vs, least three reasons for treating the Municipal Dairy as part of the city government budget rather than as a separate enterprise account. The most important is that the dairy is fiscally dependent on city government resources, both on current and. capital account. The second is that policy decisions concerning the operations of the milk scheme are taken by the municipal gpvernment , hence the city government may directly effect the distribution of income.through milk scheme subsidies. The third is that all dairy employees are considered. Msicipal Emloyees and. are subject. td the same compensation and benefit settlements. Expenditures under- the Milk Scheme have increased substantially over the past decade, as may be seen from the trends presented in Table 12. The growth in spending has been heaTiLy influenced by the growth in wage and salary rates for Municipal employees,and by expansion of operations which results in a greater employment requirement. Revenues, on the other hand, have not kept pace. There are two reasons for this. One is the provision of free milk to 40,000 school children. The other is that the price of milk has been controlled at a level below the market price. As a result of these factors, there has been an increasing deficit of the Municipal Dairy to be financed from general revenues (See Table 12). , , , , ��`. . ' t • • , ' ' � � г � ,' . г , 1�. . . ' .�1-.� _ � -- .. .. , . .^ ._ _ �.�•;г • , •л: � ((� • ' , • � i 1 � , , � ! 3 I ' . ' ' , б ,. j' . ; ' .. � , � ' ` • � • • . • . , � �11(( • � . ' • . . , ' ' . 1 ', .. .. , • ` w • • . . , . . � f . • � . . , � г . . • . . • , . � . . ,1 . ti . . } , . . � • . . " ^ � . . � � � ' • . .. . , • • • � . . . , - � . ' � • - г Р . ' • ' . • ' . . � •' � , . ' . �• ,�'�,• • 1 • , . . " ' . . i � • . • ' • , • ТаЫе 11к FIКANГIZIG REF[TSE COLLECTION � • - , ' • ' ' • ' � (i�иреев) � ' • • ' � � . , . , . �. r ' • • . " -' ' _ [, ' t�7l-;�. 1�7о-7t _ t 9f�p-7л � 4ys_с n 1�,,,т-г.н � Qf[,-�� 1 л��1-�s 1 Msh-�s� 1�з-ы� 1 чы-Ез 19ь1-fг t 9so-Ы . t •• �Сиrл,пl. fiвvcnucв . Ь2,61�718 67,20,17б 7Э,оfЭ,В48 b4,7),1Е�Ь 33,i0,pG? >,6,1Ь,2Ыу 1t?,�ь).1�б0 $L,82,92D й4�3б�ЭG1�� И5,G1,1Ь7 L1F,зL,:Ь? Llt,:ч,ЧjL •� • •. • • ' . . l в в Currunt t:z.}sendlturas 7�,sa,ze� 75,s�,б5� ч1.зг,тэз яз,о5,аз� д1,ьз,ооз 7r,1и,ч�в б�,ьо,бу2 52,4b,sa5 t,з.lс,е�� 57,ьэ,биг G�,5�,oai, зт,еб,сr�, ' �� • � 't JG1fSer�r,co -1з,2в,�8э -8,34,г+73 •1a,zз,ge5 -2е.зs,бь5 ->.и,1z.1ц1 -э1.�,б71+ -1д,z5.2чг .s,з5кч�5 •Ь.�5,5еа -1a,a2,51s .i,во,l�а ,ь,аВ,1�9о ' � . .. . • - ! � . . - - � . � . ' Mecessary Рго ert Тах Rzta to Соиег Costs ' ', � - ' � , in percent ; ' . . _ _ .• � � • • ' � ' ' �� � � ' . 51�7.}.}1 гп'•-'C3.Sti111� . . ' � � , . ' Па�епЫе цаlиа 1д.2: • 1t.,5• �.2 5•3 11•11 � �•5 5•b 11•Ь • +� • - . ° t. . WiLh Fquali..cd . , • � . ' °. - • - � ?�teatlc Va1uo , . „ УУ . P..atioa Sor . ' .. . • - Q'..Tiar-Gccup3дrs . . ' • . . snd н�n>=оrя г.4 2.6 3.i 3.2 а.е- 3.� 3•� �•7 . ' . ' , ' , ' , � ' . ' � . � i �, :` . ' • (�5 . , � 1 . . , f f 1 ' , ' . • ' ' � 1 N •� i � . . � i � . 1 �ьi . 1 +� • ' ' � ! Table 12: FINANCES AND OPERATiION OF TRE MUNICIPAL DAIRY 1972-73 1971-72 1970-71 1969-70 1968-69 1967-68 .1966-67 1965-66 1964-65 1463-64 Current Revenues 548,84,000 500,62,000 440,03,000 343,30,000 291,81,000 236,76,000 173.73,000 124,09.000 83.38.000 75,31,000 Current Expenditure, 555,94,000 504.37.000 437,48,000 344,11.000 275.94.000 239,48,000 176.27,000 125.35,000 73,67,000 67,,!1.000 (;ttrretit 5i3r1)1u.qliDefi<:it -7,10,000 -3,75,000 +2,55,000 -81,000 17,87,00.0 -2,72,000 -2.54.003 1»26,000 9.71,000 11,20,000 CapItÅl Expenditures 2.62.000 5,31,000 5,79,000 3.50,000 4,14,COO 7,08,000 5.71,(300 36.000 2.28,(-3c 5,66,~ 4 Overall St.rplus/Deflett -972,000 -9,06,000 -3,24,000 -4.31,000 +13,73,000 -9,80,000 8,25,000 +90.000 +7.43,OGO +5,44.000 1-, i na q c lm: Ilorro-,yliig 2.62,000 5,31,000 5,79,000 3,50,000 4.14,000 7,08,000 5.71,000 36,000 2,28.000 Generil Reventie of AMC 7,10,000 3,75,000 81,000 2,72,000 2>54,000 Or)zýr;ztioiiil lndicattoLis (ýuanti ty of M111c Sold 11 1,35,000 000 1,00,100 42..000 69,000 59,000 -57,000 44.000 3 1, u.)0 28,000 price 21 31 1.07 1.07 1.07 0.97 1.07 1.07 0.82 0.72 W.A.- H.A. 1/ Ilters/day. ý1 Rupeesiliter 31 In year 1973-74 fat is reduced to 3 percent. Price is 1.07 kuPeés. 41 Fata reducc-d froiMgt- 4 5 percent. Fat reduced to4 Percent. Transportation. The Ahmedabad Municipal Transport Company is effectively an entity of the Municipal Corporation. Like the dairy, it derives operating revenues with user charges but these revenues are not sufficient to cover operating costs and. debt service, and subsidy- from the city government is required. (See Table 13). From the financial data presented in Table 13, it is clear that even operating costs are not being met through operations. The explanation of this expenditure-revenue imbalance lies with rising employee compensation, the costs of maintaining an old rolling stock, and the efficiency problems which resalt from operating within a congested city; and with a fare structure which has not been changed to adequately meet these higher cost conditions. The cost increases observed for 1966-67 are in part the result of a ruling which offered AMTS employees parallel treatment with Central government employees as regards cost-of-living wage increases. There also have been substantial increases in non-labor input costs over the past decade, e.g., the results of an excise on diesel fuel. Maintenance costs are partly a functinn of the age of the equipment, and over half of AMTS buses are older than 11 years. Finally, the fact -that Ahmedabad.is a congested city with narrow streets results in a slow bus speed i and therefore a relatively inefficient operation. One result (particularly of the equipment deficiencies and the congestion) is that of 161 bus routes, only 98 meet operating costs. V The average speed of an AXTS bus is 12 Is. per hour as compared with 17 kms. in Delhi and Calcutta, according to a study carried out by the AMTS. 'r ' ' ., � � . . ,. - . ...� . . ..... . ._.. . ,. �, . _ . ...,...� ._ , .. ... _ .. _.... . � . - , •� . • . . .. . • . � . . � - ' !i • l . `j � ' � . ..� • ^� � • . yf � ,• - 'и • � � � + . . " . . - . . (. '� - ' - , .. . .. ...Э �.�• � i , , . , i., . . � ,, . . г• , . " � ._ . :1 " . . .. [ 1• -• ' ' ' - ' . .• ._.. _ . - .. .. _ .. . .. . . _ у , . . • • ..- ..,'.. • ' ' -. ....-. . ._ -_. .• _...-_. • , .. . . ,-_ - • . �� ' • ' . ' ' �. - = - ТпЫ е 7.Э : ANTS OPERATIONS ANQ FINANCINC (1л Rирееи) �, • . - !� _ . ;ь . �.�i 1572-73 1971-72 I97O-71 iЭ65-70 1968-69 1957-68 1966-67 . 7L965-бб 1954-65 lЧбз-бй ::f Curreпt Кгvепиеи Э,89,47,52б 374,56,Э88 361,06,603 305,70,608 291,Э7,Ьо8 264,81,997 231,2Э,5)3 . 2,06,15,918 1,р9,57,148 1,73,58,U11 СигтспС ехреплl[�,ги з,9а,гз,4в7 зао,оs,4г1 э54,о4,4в4 Зо4,я1,7но гsа,эе,озт 2ы,оЭ,эsг 2�c,93;6oi 2оа,7й,7sб t,7e,67,1at 1,бо,эl,ьs7 ' 4 (ееп[ s,c�ece (п .3z) 10.5х 9.1х ь.1/ 4_вх ч.sх 4.бх -• - в.1z p1.1z 9.1z " ��.;1 . _; Current surplue/nafl�lt - 8,7s,9ы - - 7,а9,озЭ 7,о2,114 - 4,)ь,13г 2,s9,s71 2,7ь,Ьйs - Э,7о,оз4 - 2,sв,аэы 1о,в5,9ь7 1з,гб,зs4 4 1 �, � 1 � ' 2/ т/ ' 4вр1саl Ехрепllture 2/ � 55,ЬЗ,о32 71,24,й25 29,79,000 3n,г7,000 22,50,000 15,00,000 15,00,000 � 35,25,0С14^ 41,74,ЬВ5 21,29�,p40 � б � 1 � Оvгг�11 Sarulue/Oeflclt - 64,38,95Э - 74,13,458 - 22,76,881 - 62,45,132 19,90,429 -12,16,020 - 18,70,ОЭч - 37,В5,842 - 36,a4,718 - р,02,й86 i i PtKA`.':Т^+',г � . . [�.Ь11с lоап В.1пеа ьу лис 2в,0о,о0о 22,ОО,о0о 1о,оо,сюо 22,о0,ооо 22;5о,ооо 15,ое,ааl 15,оо,ооо . as,0o,o0o го,оо,w0 - � 1с: TTnn.p4rL �. � 51,огс-[еги 1о.п [roo.ANC 70,94,000 9,24,425 6,50,OU0 ' 12,00,000 8,Оо,000 �o,OJ,00D 4( • . - . '; i O[F.cr lоеп■ 40,0O,OOD 3/ 1Э,29,000- 36,29,000 �;� ' � , i. � 1 'г � f � i OYPiiATION.AL INDICATU95 � ` ' � �.. - - 1iилЬсг и"с P.eeengcr п.е. 20,24,50,000 19,75,ОО,ор0 18,52,00,000 14,3Ь,00,000 I9,30,�90 1а,53,3а,000 ».а. о.и. , � (Угsгlу) . . �, �Fare рег Paa;Gn,lcr ' л.а. 17.9б 17.80 16.07 14.46 1Э.22 12.13 ,- л.а. п.а. � ' �. 1 . . (dverвge)_' � ' � Уец1у Paeeenge.r »_а. 12Э 124 120 1Э0 134 1ЭЗ .. ».w. а.в. �. рег CлpLla `4 . 4 }- .' , • 11 F.�zcludlпg DгртесlдСlоn ` 2/ Capitдl Espendlture 1е the еи� of РиЪ11г. Lодпи, рlиа SЬогt-tегш lодпr, рlии ochec lодпа. Cr,�1ta1 бхрепдl[икеи foC 1972-7Э, б5-бб, 64-65, ЬЭ-64 лга исСавl 1lgucea, ' :' che ochиra are appcuxlmeted Оу cha ebova иисл�дtiаn а! Loaaa. � , (у З/ Г.иfигс[ 5[alt Gov'[ lодп■ ' (,� ч� Сгпсааl Сеие iоеп ` � �, Э 5/ :в1ее, )QJ Рдlвс-1 Rupee. . . . . - .. . --� - .. . . д . . ,. �.__ .1.>._.., . -- . , .. . . . -,. - .. - -...__--- •• _' '" ., : � , ' , . • - - � ..-.- .• .. a'ь_т.ы�. . .,...-v гч...� :,. - � � , � , • ' 1 • The 'fare st=cture has not changed. in. several years (see Table, 14), and concessional r, te3L are offered to students.' chil:dren., 'ancr -tH6 TabLi_Tt :___'S=UIZ OF' FARES,. 197O_-o?r- Distance (Milles) Fares: Children Fares: Adults 0 - 2 0.0138 0.0275 2 - 4 0-0138 0.0481 4 - 6 0.0138 0.0550 The euxpenditure increase on transportation may be parceled between the amount of the expenditure increase due to an increase in the nmber of passengers, that dae to a rise in the price level, that dae to a rise the real expenditure per passenger-., and that due to'the interaction of all three increases.]/ The results of this analysisAndicate that expenditures increased 13.0 percent dne to the increase in passengers carried between 1967 and'1971, 13.0 pe_r'cf;n-6 because of inflatio'm* -over the same period, __p qrcent because of higher real.expenditures perpassenger and 10.2 percent, because of' tEe interaction. of all. tli=ee.. eff ecti- (See AmDerrid z I for- more detall on,' iii:s-m: i od). The o;FeiWihelm-ing'importance' of -the- Real, Expenditures Qmponent ma7 be due to =7 of a number.of factors, i.e., it may- take the fo=-- of either higher real wages and maintenance costs, or better service levels. The e=tent to which the rise in real per capita expenditure afforcLs better service can not be discerned from this partitioning. General Revenue Financed Ser7ices The grouth in current expenditure for the AXTS and the Municipal dairy enterpr-_Is--3s was 11.5 and 31.5 percent per year between'1965 and 1972 respectively, -while the growth rate current e=endit'U=e for All other Lonqral. reventie financed) ser-rices was only 7.7 percent per year over the same I/ The change in current e=enditure might be viewed as a total differential of current exzenditi3=e with resmect to e=endituze per passenger, the number of passengers, and inflation, and these are denoted as E p , S, amd I res-pectivel-7. Also let E denote current expenditure on transportation. Notationall , the differential is: dE (EP + dE ) (S + dS) (I + dl) (EOSI) .... .. . ... - . .... .. 3T. period (see Table 15). The composition of this growth is presented in Table 16 below. To the extent that the Transport and Dairy eterprises operate in deficit, and use the general revenue of the Abnedabad Municipal Corpora- tion to offset this deficit, they leave less revenue to be spent on the ser- vices financed from general revenue. The deficit of the current account of the Dairy in 1972 was 3.75 lakh IMpees and the deficit in the current. account of the AMTS was 7.89 lakh Rupees. However, the purpose of the dairy undertaing is to distribute milk to families who could not afford to bT milk from other sources, hence the deficit might be viewed as a result of an. intentional redistri- butive measure. Still, the extent to which the scheme actually is redistri- ive depends on the extent to which expenditures' on milk weigh more heavily on the incomes of the- poor than do the mainstays of the general account-property taes and actroi payments- The AMTS finances its deficit not from. the general revenue of the AMC but from. borrowing from. the AMC, the Central Bank and Gujarat State, so'it does not drain general revenues from the AMC. owever, the fact that bond funds are limited means that there is a cost to the AMC, i.e., there is an opportunity cost of the funds loaned to the ATS. V. Revenue Structure and Trend The structure of revenues for consolidated local governments in the Ahmedabad Urban Area (described in Table 7 above) shows a heavy dependence on general taxation and on user charge financing, and very little reliance on external assistance. The primary user charges - bus fares and milk receipts - hav-4already been discussed above, therefore, the discussion below will center around the major forms of taxation, grants, non-tax general revenues, and borrowing. 38. Table 15: CURT EENDITURE TRENDS: 1964-1972 (in lakh Rupees) General Revenue Financed Milk AMTS and Year current expenditures A.1. T. S Scheme Milk Scheme 1964-65 674.57 178.67 73.62 252.36 1965-66 767.15 2c8.74 125.35 334.09 1966-67 797.92 234.93 176.27 411.20 1967-68 884.20 262.03 239.48 501.51 1968-69 965.64 288.78 275.94 56.72 1969-70 975.90 309.87 344.11 653.98 (' 1970-71 1,058.78 354.04 437.8 791.52 1971-72 1,130.47 382.45 50.37 886.82 Average annual 7.7 11.5 31.5 19.7 percent Increase 39 Table 16: PER CAPITA EXPENDITURE BY FUNCTIONS FOR AMEDABAD MUNICIPAL CORPORATION 1965-1972 (In -Rupe es) __ __ 1965 1966 19 7 1968 1969 1970 1971 1072 Water Suplv 6.21 5.36 4.00 5.72 4.87 8.0h 8 ý 08 5.98 Currer 2.46 2.68 2.31 3.01 2.97 3.23 3,64 3.80 Caital 3.75 2.68 1.69 2.71 1.90 4.81 4.44 2.19 $2:erage & Drainago l.99 5.25 6.77 13.38 12.07 7.92 6.50 5.97 Current 2.47 2.40 2.18 2.16 . 2,17 2.05 3,10 3.20 Cital 2.52 2. 5 4.59 11.22 9.91 5.87 3.40 2.7 Roads & Žridges 7.44 6.89 4.60 5.0, 4.97 4.79 6.31 9.06 Crrent 3.50 3.93 3.22 3.69 3.71 3.47 3.86' 4.32 Cita1 -3.94 2.95 1.37 1.35 1.27 1.32 2.94 4.74 .,ealth 8.90 10.05 10.37 10.27 10.84 10.49 11.30 11.74 Current 8.39 10.05 10.37 10.27 10.84 10.49 11.08 11.74 Capitel .50 n.a. .21 Education~ 14.81 19.01 20.17 21.38 23.52 23.53 23.59 26.14 Ourrent 12.83 17.31 19.28 20.17 22.80 23.13 22.81 24.67 Capital 1.98 1.70 .89 1.21- .72 .41 .78 1.46 Welfare .29 .57 .-0 .83 .84 .58 .9 .99 Current .29 .57 .40 . .83 .84 .58 59 - .99 Capital --- --- --- --- --- --- --- --- Fire .70 2.27 1.31 1.15 .89 .88 1.58 2.38 Current .49 .65 .63 .65 .71 .67 .87 .90 Capital .21 1.62 .68 .50 .18 .21 .72 1.48 Refuse Collection 4.01 4.86 5.00 5.64 6.24 '5.93 4.76 4.62 Current 4.01 4.86 5.00 5.64 6.24 5.93 4.76 4.62 Caoital --- General Adminis- tration, 5.70 6.55 6.95 7.66 8.54 7.64 8.24 8.36 Current 5.34 6.55 6.95 7.34 8.53 7.63 8.12 8.36 CaDital .35 - .31 .01 .01 .12 -- 1/ AMC expenditures plus amt sciLf-financed 40.- Tabie 16: _PER CAPITA EPENITURES BY FUNCTIONS FOR AHNMABAD MUNICIPAL CORPORATION 1965-1972 (In Rupees) (cont) 1965 1966 1967 1968 1969 1970 1971 1972 Housing 5.66 4.33 2.88 .96 .77 .22 1.11 2.20 Currcnt .19 .2 .19 .01 ----- --- Capitai 5.47 4.04 2.68 .95 .77 .22 1.11 2.20 Tr,nsortaticn 13.74 15i13 16.81 18.15 19.37 20.12 22.29 23.27 Current 13.74 15.43 16.81 18.15 19.37 20.12 22.29 23.27 Capital -- - --- --- Debt Service 11.80 12.93 12.85 13.77 13.-7 14.06 15.33 1i.i86 Current 11.80 12.93 12.85 13.77 13.87 ij06 15.33 1h.c6 Capital --- -- --- --- - --- --- --- Land Redeveloment .01 .09 .L,9 .83 1 .94 1 .12 .17 Current --- --- --- --- --- --- --- --- CapDital .01 --- .09 .49 .83 1.94 1.12 .47 (Milk Scheme 5.-80 9.29 13.02 1.7.08 18.79 22.57 27.91 31.01 Current 5.63 9.27 12.61 16.59 18.51 22.34 27.54 30.69 Capital .17 .03 .41 .49 .28 .23 .36 .32 arke ts .27 .17 .15 .17 .17 .16 .18 .19 Current .15 .17 .15 .17 -17 .16 .17 .19 Caoital .12 --- --- --- --- .01 --- Recreation 1.80 1.72 1.37 1.43. 1.42 1.59 1.73 1.82 Current 1.29 1.148 1.30 1.35 1.39· 1.37 1.145 1.50 Capital .50 .24 .08 .08 .04 .23 .28 .32 Other 8.23 5,99 .85 1.40 3.00 1.81 2.61 2.11 Current 3.65 .25 .25 .29 .40 .47 .52 .43 CaDital 4.58 5.74 .60 1.11 2.60 1.35 2.08 1.67 TOTAL 100.25 ro.67 107.47 124.53 131.01 132.28 143.37 151.79 Current 76.14 88.82 94.37 104.11 U2.50 n5.69 126.07 133.54 Capital 24.11 21.8,5 13.10 20.142 18.51 16.59 17.30 18.25 1/* 41. Taxation / Tax revenues have ranged between 30 and 40 percent'of Ahmedabad local goverrment resources over the past decade (see Table 7). However, on a per capita basis, tax revenues have grown at a rate of about 6.2 per- cent per year between 1964 and 1972. Over the same period, the price level increased at a rate of 5.7 percent, hence real per capita tax revenues grew at a rate of 0.5 percent (see Table 18). This buoyancy in receipts has come about primarily because of the revenue growth of the Octroi, which has increased as a share of tax revenues from 36 percent in 1964 to 52 per- cent in 1972 (see Table 17). The property tax has been less dyanic over this period, and fell from 61 to 43 percent of the total. All other local taxes together account for only about 5 percent of tax receipts. The increase in total tax revenue in Ahmedabad between 1964 and 1972 may be viewed in terms of the increase in population, prices, and real per capita tax revenues,- as in the analyses above for transportation and total expenditures. Using the same partitioning approach as presented above, the analysis indicates that 26.8 percent of the increase in tax revenue may be allocated to increases in population, 51.1 percent to inflation, 3.5 percent to increases in real per capita tax revenues, and 18.6 percent to the interaction of all three effects. 1/ Taxation is taken here to includ- the water and qonservancy surcharges on the property tax. 2/ The change in total revenue (R) may be expressed as the total differential of total revenue with respect to population, inflation, and per capita total revenue. Letting the above three variables be represented by F, I, and Ro respectively; the total differential can be written: dR (P + dP) (RO + dR ) (I + dl) -(PIE-) p Table 17: TAX RKENUE STRUCTURE AND TRENDS (Rupees) 1971-72 1970-71 1969-70 1968-69 1967-68 1966-67 1965-66 196i-65 963-Ch Per Capita Total G 1neral Taxes 60.48 55.22 57.59 57.30 . h0.41 42.42 37.47 39.02 37.27 Pk,rc -t;.:r Distribution: Gerieral Property Tax 33 35 39 ho 26 39 43 43 h. hater Property Tax h 5 5 6 4 5 6 6 7 Conservýncy Property Tax 6 7 8. 7 6 7 , 9 10 10 :Atroi 52 48 h3 43 61 46 ho 37 36 Vehicie Tay- .2 2 2 3 3 3 3 3 3 Other Taxes 3 3 3 1 1 1 …― 44. It is clear from these results that real per capita taxation has increased only slightly between 1964 and 1972, however, a more relevant issue is whether real per capita income has grown faster than real per capita tax revenue. This would imiol that tax effort in Ahmedabad -has declined between 1964 and 1972. Unfortunately, income estimates are not available for Ahmedabad. However, income figures are available for Gujarat State. In general, we expect the per capita income, and the growth of per capita income in the major cities to be higher than the state average.' Thus the income figures and the growth of per capita income in Gujarat State may be viewed as a lower bound to the level and growth rate of per capita income in Ahmedabad (see Appendix Table 1). Since real per capita income in Gujarat State increased 8.5 per- cent between 1964 and 1971 (see Appendix A) and real per caDita tax revenues of the AMC increased by only 3.8 percent between .1964 and 1972 (calculated from Table 18), the tax effort of Ahmedabad apparently has decreased ov6r the 1964-1972 period.l/ Property Tax Legal Structure. The property tax in Ahmedabad has three co=onentsi a general tax which ranges in rate from 12 to 30 percent of assessed annual rental value, a water tax of 2.5 percent, and a co4servanTy tax, of 3 Percent. There is also a state education cess-which is levied on the same base, but ranges in rate from 3 to 10.5 percent depending an property class. The normal orater and conservancy property tax rate does not include "special properties" -- factories) mills, the Ahmedabad Electric Company, hotels, boarding houses, restaurants and cinemas pay a differentially higher charge (see Table 19). 1/ E*7en this may overstate the revenue elasticit7 since there has been no effort here to "clean" the revenne increments by removing the effects of discretionary actions. h5. The base of the property tax is gross a=nual rent, and the liability for payment is on the property owner. If a. residential property is not owner-occupied, the base used is either actual rent receipts or estimated "faix" rent, whichever is greater. For t,iwner-occupied housing, an annu.al rent is imputed on the basis of the physical construct of the home. Fur non-residential properties, either a profit or a construction cost method of assessing annual value is used. The value of machinery and equipment is not taxable unless it is a part of the structure. Most of the older properties in Ahmedabad are subject to a rent control (standard rents are based on 1940 levels) but assessment by the AMC is still made on a basis of gross annual rental., 7,alue. Recently there was a Supreme Court judgment which would require I; AMC to assess these properties at controlled rents, rather than at estimated market rents. However, the AMC has introduced an amendment to ign;ore rent controls in; assessing properties, and.a judgment on this issue is now pending. The implications of the rent control issue are seriocs since it is estimated that more than 80 percent of all rented properties in the £50 are under the rent control. Assessment Procedure. The basic methods of assessment used by the AMC are comparative rents for non-residential properties. With respect to non-owner occupied (rented) properties, assessmet(t Ls based on rents actually realized by the landlord, if such rents approximate a fair market rent. Both the landlord and the tenant are required to produce a rent 46. Table 19: PROPERTY TAX RATE STRUCTURE Genera! -ducation ;äater Conservancy Total oal Vrl 2r .ate Oess T Rs 0-100 12' Exempt 2.5 3.0 17.5 101-200 12 Exempt 2.5 3.0 1 7.5 201-300 12 rxempt 2.5 3.0 17.5 301-500 12 3 .2.5 3.0 20.5 501-i0co 15 3 2.5 3.0 23.5 1001-2000 22 5 2.5 3.0 32.5 2001-2500 23 5 2..5 3.0 33.5 25 01--3C00 23 ö 2.5 3.0 34.5 30014,500 28 6 2.5 3.0 39.5 45)i-5xo 28 7 2.5 3.0 ' o .5 o500 30 7 .2. 3.0 h2.5 Non Residcntial 0-300 30 Exempt 101) 9 49.0 301-1000 30 4.5 10 9 53.5 1001-2500 30 7.5 10 9 56.5 250-4500 30 9.0 10 9 58.0 over-4500 30 10.5 10 9 59.5 ) Being contested in the Courts in favor_f th. standar.. water charge 47. payment receipt. If the assessor feels that the stated rent is not a fair one, the estimated average market rent for the neighborhood is used. This neighborhood average is estimated for a sample of properties on which true market rent data are available, with the judgment of the assessor playing a major role in the combining of these data to reach an average which is applicable to any given property. Finally, a significant proportion of city properties is subject to rent controls, but these are assessed strictly on a basis of estimated market rents. Owner-occupied residential properties are assessed on a different basis. Among the important considerations in determining assessed value are the location of property within the City's 57 wards, the specific amenities of the property, construction material, ventilation, and carpet area. There are graduated assessment rates depending on*these considera- tions, but again, the judgment of the assessor plays a major role. Though there is no assessment manual to which assessors strictly adhere, the range of assessment rates shown in Table 20 is used as a guide. The determina- tion of the exact assessment within any given class is determined by the assessor on a basis of the amenity and location considerations outlined above. The standards presented in Table 20 have not been adjusted since 1966-67. At that time, the limits in each class were raised by Rs. 1.2 per square meter, i.e., the previous ranges were 10.8 to 16.8, 6.0 to 10.8, and 3.6 to 6.0. The lowest of these classes roughly approximates the asses- sment of hutment dwellings. On a basis of comparison of assessed values of similar structures, rented and owner occupied, it is argued that the result of this assessment procedure is a substantial preferential assessment. 48. of owner-occupied properties, i.e., assessment at about one-fifth that for non owner-occupied properties. Since owner occupiers are likely to be in higher income classes than renters, cet. par., this assessment bias may heighten the regressivity of the property tax system. The rationale of this feature of the tax is difficult to understand, since the property tax rate structure is graduated, apparently to increase the tax burden on higher income residents. The valuation of commercial and industrial properties is usually made on a basis of the cost of reproducing the physical facility. 49. Table 20: ASSESSMENT STANDARDS FOR OWNER-0CCUIED HOUSING Annui Assssed EstimtedJbilding VE.lUC p0r sqlare, Dscr,ripni3*.o n L:.cMetor (in Ps) Eetter aildIngs Reinforc-d Over 80 -ears 12 - 18 concro e Mediurm Buildigs izonry and $0-OQ years 7.2 - 12 wood con- strucdicn; tile roo Semi-Pcca ilIdings Not pe.anent Under C0 ycars 4..8 - 7.2 conitruc onw includecs che ls 50. Most important of the "special properties" are the mills and factories (primarily textile manufacturing) which account for about half of all employment in the Ahmedabad metropolitan area. Before 1966-67, the valuation base was land value at the time of purchase and building value at the time of erection. These were translated to gross annual rental value by using capitalization rates of 6 and 7 percent for the land and buildings respectively, and the result was a constant property tax yield of 45 lakhs. However, in 1967-68, the assessment base was changed by the AMC from the original value of land and buildings to the assessed market value, and a uniform capitalization rate of 6 percent was adopted. The result was an increase in tax liability from 45 lakhs to 1 crore 50 lakhs. The mills and factories took the case to court and refused to pay the tax during the period 1967-68 to 1971-72. In the latter year, negotiation between the AMC and the mills resulted in the mills paying annual property-taxes of an amount equivalent to a 25 percent increase over what they were paying in 1966-67. Some 52 of the 67 firms accepted the agreement and, therefore, property tax receipts in 1970-71 reflect these three years of back payments. Still, 15 mills have not agreed to the settlement and there is a total of 2 crores of property tax revenue outstanding. Small commercial establishments are usually assessed on a basis of estimated fair rent for the premise, even if the owner also resides there. Where construction cost has been used in the assessment, the methods used and resulting controversy is exactly as described for the mills and factories above. The assessment of the 26 old and 6 new cinemas in Ahmedabad had resulted in property tax payments of four lakhs before 1970-71 . The assessment base was the lease amount in cases where cinemas were leased, 51.. and annual profits where cinemas were owned. 'In 1971-72, the assessment base was changed to the market value of land and buildings with a capitalization rate of 6 percent, with the result that the property tax demand has risen by 88 perceint to 7.5 lakhs. The.cinemas have refused to pay the increment and a challenge of this assessment method is now in court. TQtal annual profits of the Ahmedabad Electric Company were taken to be equivalent to gross rateable value, with annual revisions, before 1966-67. However, in the latter year, the profit method was ruled illegal and eventually it was ordered that the constructinn cost.-method be used. In 1968-69, the Ahmedabad Electric Company agreed to pay 22 lakhs per year while the court case was pending, and did the same in 1969-70. In 1970-71, when the construction cost method was applied, it was determined that the total tax liability was 15 lakhs. However, the Electric Company challenged that assessment and paid an amount of only 5 lakhs for 1970-71, 1971-72 and 1972-73. The matter is still in court. Buildings of the Gujarat State government and the Indian Union government are effectively subject to property taxation in that they Make a centribution in lieu of the rates. 1/ The 5,000 state government properties pay a total tax of 6 lakhs and the 13 Federal government units pay a tax of 1 lakh. Reassessment on these properties is carried out every five years by an appointed, neutral special assessor. The assessment is made on a basis of the original land and construction cost, and a capitalization rate of 9 percent is used. Finall ,.nly 75 percent of the estimated general property tax is paid,and these properties are excluded from the water and conservancy charges. With respect to .AMC rental housing, the property tax is collected. as a component of the rent. 1/ Technically the constitution of India provides that union properties will be exempted from all taxation by States. 52. Administration. The total administrative cost of the property tax was 18.6 lakhs in 1972 for collection and administration , or about 4 percent of revenues. Total manpower involved is 450, of which only 5 are assessors. None of the five are,formally trained valuers. Given that there are 77,400 propeTties per assessor, the annual updating of property valuations is not possible. However, a blanket revaluation is undertaken every four years with the help of a number of State government valuers. Stringent methods were employed in 1969-70 to increase the efficiency of property tax collections. About 500 properties were auctioned and about 3,000 water connections were cut off for failure to pay taxes. The total number of tax appeals is now down to 4,000 and the total out- standing tax involved is only 10 lakhs. Revenue Yield. Property tax revenues in Ahmedabad over the past decade have grown at a rate which is less than that for city population. There are a number of reasons for this relatkvely slow growth: (1) Non-residential assessments have been challenged (successfully) and property tax payments have not been made during the litigation. (2) Preferential assessments of owner occupied housing has robbed the city of much of the natural growth in assessed value. (3) The property tax base in general has not grown rapidly, particularly in recent years. The first of these allegations has been dealt with above in the description of the assessment challenges before the courts. The nature of 1/ Housing units with kitchens. 53. the preferential assessment of owner-occupied housing has also been described above--the city assessor estimates that owner-occupied units are assessed at about one-fifth the level for renter occupied units. If this esti- mate is realistic, then dropping the preferential assessment would raise rateable value by an amount equal to five times the present rateable value of owner occupied properties. A conservative estimate might be made. Approodmately 23 percent of assessed property units are owner-occupied. Assume that these units also account for 23 percent of (residential) net rateable value. With this assumption, and an assumed four-fifths under- assessment, the removal of the owner-occupier preferential assessment would result in an increase in the property tax base (in 1973) of 75 percent. This rough estimate suggests the potentially enormous opportunity costs of such preferential assessment. The property tax base has not experienced rapid growth. Growth in rateable value, in theory, should result from one of two factors-- a natural growth in rent levels and/or property values, and new constrtiction. The latter usually is added inediately to the property tax rolls, while the former requires revaluation. The AMC has the practice of reassessing pro- perty every four years, but very little reassessment is carried out in the intervening years, therefore, the property tax growth between revaluation periods is primarily due to addition of newly construction units to the rolls. The historical growth in net rateable value is presented in Table 21, with the revaluation years starred. A number of trends are apparent. The first is that the percent of net rateable value in non-residential pro- perty has declined from betwe-n 1965 and 1973. Also note that the net rate- able value of property in Abmedabad grew at an annual rate of only 5.7 per- cent, while population in Atnedabad was growing at an annual rate of 3.3 54.. percent, thus the per capita net rateable value of property grew at an an- nual rate of 2.4 percent over the period 1965 to 1973.. Property tax rev- enues (c.f. Table 22) grew at an a=nual rate of 5.4 percent, and per capita property tax revenues grew at an annual rate of 2.1 percent. Elasticity The rate elasticity of the propety tax is defined here as the ratio of the percentage change in property tax revenues to the percentage change in the net rateable value of property, The rate elasticity ignores the question of how rateable value (the tax base) responds to growth in the local economy. If the value of the rate elasticity is unity, then tax revenues increase in the same pro- portion as the increase in net rateable value of property; if the value of the rate elasticity is greater than unity, then tax revenues increase more than'proportionally to an increase in net rateable value) and if the value of the rate'elasticity is less than unity, then tax revenues respond less than proportionally to an increase in net rateable value. Since the rate sructure of Ahmnedabad's property tax is progressive, the rate elasticity of the tax will depend on which property value classes grow more rapidly. If the gro-4h in net rateable value occurs in properties taxed at the lowest rates, i.e. in rate classes below the average, then the rate elasticity of property tax will be less than unity. Similarly, if the growth in net rateable value occurs in properties taxed at above average tax rates, the rate elasticity of the property tax will be greater than unity. If all rateable value groups increase in rateable value by the same percentage, then the elasticity of the property tax is unity. ss. Table 21: GROWTH IN PROPRTY TAX BASE Net latcable IUt Ratable Valuc per Percent of 'ot Ratciblc Valucoy u Valuc 1s5intial Noesdetil , ResidOntialc, Non-t Jiincntial 1965 11,hc,6h,i62 283.3 1,16,605.7 75.O 25.c 1966 12,23,13,845 300.2 1,25,051.9 73.4 26.å 1967*- 14,10,39,554 363.3 1,03,760.2 79.3 20.7 1968 18,26,24,650 363.0 2,29,532.4 6I.7 35.3 1969 17,4o,30,547 380.0 1,72,811.3 72.6 27.4 1970 17,69,20,550 380.6 1,57,976.o 72.2 27.8 t971 16,87,02,035 385.5 n.a. 80.7 19.3 1972* 18,01,27,752 380.6 30,296.5 81.3 18.7 1973 18,59,73,902 387.6 31,236.Q: 82.8 19.3 Average annual. rate of growth 5.7 1/ 3t Rateable Valu avaJIÅle for 1967 or 1971. It w estin. by deducting 10% of Cro.- R.tcable Value fron G:oss Rateable Valtue, because there is a 1'o deduction for landlcrJ for maintenance ot properties from the Grosa Ptej ble Vaue. 2/ The decline shown here is due to a marked increase in the number of non-residential properties assessed,, and a decline in rateable value (due to litigation). 56. Table 22: PROPERTY TAX REVENUE TRENDS (in Rupees) Total Property Revenues per Rupee Tax Revenues 1/ of Rateable Value 1971-72 -446,04,620 0.248 1970-71 -423,29,674 0.251 1969-70 .476,46,123 0.269 1968-69 457,48,263 0.263 1967-68 206,24,381 0.113 1966-67 305,53,781 0.217 • 1965-66 297.77,515 0.243 1-6h-65 304,53,447 0.267 Average annuail rate of growth 5.4% 1/ Total Property Tax Revenues represents the sum of the Water Tax, General Tax, and Conservancy Tax. 57. The average property tax rate including General Tax., 'Water Tax, and the Conservancy Tax is approxi.mately 0.25 (Cf. Table 22). Residential propetty in ra.teable value groups 1000 Rupees and above have tax rates that exceed the average of 0.25 (CF. Table 19). Since most of the growth in rateable value of property has occurred in rateable value groups above 1000 Rupees between 1969 mid 1973, and since most of the rateable value of property is in these Value Groups (OF. Table 22), it might be expected that the rate of elasticity of' the residential property tax in Ahmedabad. would-,be greater than ur,L~ On the other hand, the rateable value for special properties has not increased rapidly, and the tax payments per Rupee of assessed. value has declined because -of the litigation described above. The rate elasticity calculated here is an are elasticity,_bec.Ause there are insufficient degrees of freedom to "estimate. a pointF last_icit y using regression analyais. 'The formtula for the rate arc) e laticity omr) between 1972 and 1971 is: 2 a- R 2t-V TirR7 7 V72 + 71 (6) where aR =the change in property tax revenuae between 1972 and 1971. R72, R?l = property tax revenu~e in 1972 and 1971 respectively. &V=the change in net rateable value of property between 1972 and 1971. 772, 771 = net rateable value of propertby in 1972 and. 1971 respectively.- The arc elasticity between 1971 and 1972 is 0.80. This result wrould seem to bear out a contention that the litigation problems with non- residential properties have pulled the rate elasticity below unity. 58. However, there clearly are problems with this estimation, and may cause this estimated value to be questioned.! These qualifications notwithstanding, the are elasticity estimated in this sector and the lower growth rate of net rateable value both indicate that the rate elasticity of the property tax is less than unity. With some idea of the rate elasticity, the subsequent question is how does the base of the property tax respond to population growth, i.e., the mag- nitude of the base elasticity. (nb) An estimate of the elasticity of net rateable value of property with respect to population is estimated here from the following relationship: V7 F (7) where V = net rateabl vlue of property for years 1965-1972. P = population of AMC for years 1965-1972. a = the (constant) elasticity of net rateable value of property with respect to population. Estimating equation (7) in logarithms (See Table 23 for data) using ordinary least squares, yields the following results: log (7) -L.23 + 2.01 log (P) R= .72 (8) (3.8) 1/ Two of which are particulary important. The first, is that the net rateable value for 1971 is estimated as a constant proportion of gross rateable value for 1971 and, second, tax revenues for 1971 include three years of back taxes owed by the mills and factories, which did not pay taxes during the litigation of assessment procedures from 1968 through 1971. in fact, any estimate of (ir) for the period 1968 through 1972 is likely to be misleading because of the errat- ic payments by the factories-and mills. One could obtain an elasticity esti- mate between years 1967 and 1966j but there was a change in the tax rate in 1967, so that this estimate of the' elasticity is also not correct because it is based on the old rate structure. 59 Table 23: INCREASES I NET RATEABLE VALUE FOR PR)PERTY VALUE CASS - 1969 and 1973 Residential Residential Value Group Net Rateablevue 1969 Net Ratable Value 1973 Percent decrease (In Rupees) (In Rupees) 1-100 81,53,036 99,73,705 22.3 101-200 143,281.323 1,50,59,034 5.1 201-500 281,35,567 3,07,94,846 7.2 501-1000 227,51,267 2,75,47,078 21.1 1001.-2000 16,959,960 1,95,56,717 24.9 2001-3000 87,01,032 1,11,27,465 27.9 3001-5000 93,86,768 1,21,17,970 29.1 5001-above 186,18,685 2,41,66,693 29.8 Special Properties Non-Residential Value Group Nc,t Rateable Value 1969 Net Rateable Value 1973 Percent decrease 1001-2000 1;890 1,890 - 2001-3000 7,722 19,110 147.5 3001-5000 15,8"h 69,662 338.7 5GO1-above 4,76,70,453 3,55,39,732 -25.4 60. where the figure in parenthesis is the t-statistic and, indicates that the elasticity estimate is significantly different from zero at the 0.01 level of significance. If, on average, the elasticity of the net rateable value of property with respect to population is 2.0, and the arc elasticity between property tax revenue and net rateable value of property is 0.8 between 1971 and 1972, then the overall elasticity of tax revenues with respect to population between 1971 and 1972 is 1.60. Since the elasticity of the property tax is less than both the elasticity of current expenditure and revenue from own sources with respect to population, it may be concluded that revenues from the property tax are a shrinking proportion of revenue from own sources and of current expenditure. More specifically, a one percent in- crease in population tends to raise current expenditures and revenues raised from own sources by approximately 2.9 percent, by property tax revenues by only 1.6 percent. Again, if we take the average annual rate of growth of per capita income in Gujarat State as a lower limit on the growth rate in Ahmedabad per capita income, we would conclude that the property tax is highly inelastic. Specifically we would conclude that a one percent increase in per capita income would produce a 0.21 Dercent increase in property tax revenues. 60a. Octroi Legal Structure. The most important local Lax source in Ahmedabad is the octroi, a form of sales tax collected on goods entering the city. It is levied on both an ad valorem and a specific basis. Four classes of goods are covered.under the octroi: food, cloth and yarn, industrial goods and machinery, building materials, and provisions and consumption articles. Goods which are exempted are: (a) Government purchases (b) gold, silver and jewelry (c) khadi (spun-woven cloth) (d) green fruits and vegetables (e) milk (f) salt (g) arms and ammunition (h) railway goods. Goods in transit are subject to a small supervision or escort fee. Ad valorem rates are generally levied on finished goods, and specific rates on intermediate goods. There are 55 rate classes in total. Rates were revised substantial in 166-1967. 61. Administration. The tax is collected at 33 octroi stations -- I 18 rail,15 road, and 1 air. When the levy is specific, there .is usually no dispute in the assessment; however, with respect to finished goods where it is levied on value, an invoice is required and must be examined at the octroi station. There is a market value manual which is used by assessors to double-check the invoiced amount. Revenue Yield. The growth in octroi collections over the past decade has exceeded that of all other locally raised taxes, the Gross Revenues have grown- to about 601 percent of total tax yield. Its annual rate f of growth of 10.3: percent per capita far exceeds the annual growth in per capita income (see Table 24). Over half of gross collections are due to.machinery and industrial. goods, and the share of this tax class has increased slightly over the-past decade. It is especially notable that the food group has declined in impcrtance, and this appears to be largely due to the rate structure reforms in 1966-67.' Estimation of the income elasticity of octroi revenues is not possible because of the absence of income data. However, it is possible to examine the growth in octroi revenues with respect to population increase or simply over time. But since the focus of such analysis should be on the natural or autcmatic revenue growth, revenue increases due to discretionary 62. changes (e.g. rate changes) ought to be removed. The complicated rate structure makes a sophisticated "cleaning" of the tax yields impossible,i/ but since these rate changes took place in one year, 1966-67, their effect might be captured by comparing yields before and after that period. Consider the following general model: Rt - R1t + R2t +* *, + nt (9) where Rit = revenues from the first class of goods in the t - year. The object is to estimate the marginal rate of revenue growth per year, dRt from the equation dt Rt= + t (10) where a is the marginal rate. In order to take account of the discretionary change in the tax rate in 1966-67, a dumy variable is intro- duced which has a value of 1 in years after the change, and 0 before. Now equation (10) may be reestimated from R. = a + 01 t B2Zt + 03 (tZ) (11) U where Rt = octroi revenues t = time, 1962-1971 Z = du=y variable; Z = 0 for years 1962-1966 and Z = 1 for year 1967-1971 A least squares estimation of the parameters in equation (11) yields the following equation: Rt = 14070 + 1736 t - 9428 Zt + 2889 (tzt) R2= .988 (12) Effectively there are two equations: one for the years preceding the 1/ For a discussion of the forecasting problem, see Roy Bahl "Tax ReTenue Forecasting in Developing Countries" International Monetary Fund Departmental Paper, O'ctober 1973. 63. discretionary change (1967), where Z = zero, and another for the years following the discretionary change where Z equals one. The equation for the period preceding 1967 is: Rt = 14070 + 1736t (13) the average annual growth in tax revenues for the 1962-1966 period is 1736 rupees per year. The equation for the period following the discretionary change, 1967-1971, is: Rt = 4642 + 4625t (14) the average annual growth in octroi revenues for the period 1967-1971 is 4625 rupees per year.1/ Thus, on average, the discretinnary change in 1967 increased marginal octroi revenues by 2889 rupees.per year. It may also be useful to determine how the discretionary change affected octrai revenues on a per capita basis.- Accordingly, the following equation is estimated: Rp + t + +2t +3(tZt) (15) where R = per capita octroi revenues, and the other variables are as above. A least squares estimation of the parameters of equation (15) yields the following equation: Note that the slope of the line for revenues with respect to time increased from equation (13) to (14), but the intercept declined. This result is expected because the point where the slope is al- lowed to change is fixed at 1967, or t = 6, and at a specific point if the slope of the line decreases the intercept must rise. Specifically the dummy variable technique forces this result. 64. R = 12.53 + .8700t - 2.67Zt + .1370 (tZt) 2 = .979 (16) Again there are two equations: one for the years preceding the discretionary change and another for the years following the discretionary change. The equation for the period preceding 1967 is: R = 12.53 + 0.87t (17) p The average annual growth in octroi revenues per capita for the period 1962-1966 is 0.87 rupees per year. The equation for the period following the discretionary change, 1967-1971, is: R.p = 9.86 + 1.007t (18) The average annual growth in octroi revenues per capita for the period 1967-1971 is 1.007 rupees per person per year. Thus, on average, the discretionary increase in 1967 increased marginal per capita octroi revenue by 0.137 repees per year. Other Taxes A number of other taxes are levied by the AMC, but together these account for less than 5 percent of tax revenues and less than 1 percent of total revenues. The largest of these is the vehicle tax, which is levied on vehicles registered to Ahmedabad addresses. The annual tax varies with type of vehicle: motor cars (108 Rs-,), motorcycles .(Rs.54), buses (Rs.30Q), and trucks (Rs. 500). The only other'local tax with a noticeable revenue yield is the theater tax. The tax is levied on the proprietor at Rs.15 per show per day, i.e.,Rs,60 per day per theater. If the movie is not first run, the rate is only Rs.5 per show per day. Table 211: OCTROI REVEn5E Bt CLASS OF GOODS Total Net Octroi * Revenue as a Percen age Distribution of Gross Receipts 1 Percent2 of.,Total Cloth &--Yarn Building Provision and Machinery Motors Other Fiscal Year Amount Iercapita AMC Revenues Food Group OcoIs Materials Articles of Consumption and Industrial Materials 1962 158082711 13.3 37.2 20.3% 12.9 8.3 7.6 47.3 3.6 (12.5%)3 (9.0%)3 1963 17777831 1t.5 38.8 18.5% 15.4 8.5 7.9 146.0 3.8 (7.4%) (0.1%) 1964 19087812 15.1 36.5 17.6% 15.4 9.7 7.8 4,2 4.2 (8.6%) (4.6%) 1965 20734928 15.8 36.7 16.0% 16.0 9.4 7.1. 47.6 3.9 (11.0%) (7.6%) 1966 23010025 ' 17.0 40.4 15.7% 15.3 8.6 8.5 48.0 3*9 (33.2%) (28.8%) 1967 30651627 21.9 49.6 11.b% 17.3 8.7 10.2 W.-6 4.6 (28.6%) (24.7%) 1968 39h16090 27.3 . 60.8 9.8% 13.5 8.3 8.4 49.9 5.2 (7.3%) (11.0%) 1969 42292860 28.4 13.4 10.4% 1i.6 8.5 7.7 51.> 4.4. (5.9%) (2.5%) 1970 W1805701 29.1 h3,1 9.9% 15.3 10.7 7.7 51.5 5.0 (14.0%) (10.7%) 1971 51087334 32.2 I48.4 10.8% 1(.2 8.0 7.9 51.0 5.1 Average Annual 13.9% 10.3% Growth rate of Collections. 1. Gross Revenues 2. Net Octroi Rev,no3 as percent of total revenue; Net Octroi Revenue-Gross less refunds. 3. Percent increaseis between years shown in parenthesis. I .Ar Al-t 66. VI. Intergovernmental Assistance The three major local government bodies in Ahmedabad --the AMC, the School board, and the AMTS-- receive direct Central Government Assistance only for the AMC slum clearance, nursing, and family planning programs. There are, however, seventeen categories of Gujarat State government grant-in-aid assistance to local governments. In most cases, ,the amount to be distributed and the distribution formulae within the State are not objectively defined but are rather the result of the State Government budget process. General Purpose Grants Education Cess. There are three general assistance grants-in- aid to the Ahmedabad Municipal Corporation, which together account for less than one-fifth of total state government grants. In reality, all are shared taxes. The most important'general purpose grant is the education cess, a surcharge on the local property tax, which is levied by the state government under the Education Cess Act of 1962. The tax is collected by the AMC, and one-third of the total is retained for general use.]/ The base of the education cess is the same as that for the general property tax, and the rate is a graduated percent of net rateable value with differential rates as between residential and non-residential property. The rate structure of the education cess is described in Table 25 below. Other Less important sources of general purpose assistance are shared land tax revenues and grants under the Motor Vehicles Act. With respect It costs the AMC approximately two lakhs to collect the cess, and the grant is presently about 18 lakhs. 67. to the former, the state government collects land revenue on open lands and property tax on other lands used for non-agricultural purposes - particularly in town planning areas. It is argued that because the growth in value of these lands is partly due to activities of the AMC, the revenues accruing from these sources ought to be shared with the AMC. Accordingly, ?5 percent of the non-agricultrual property tax collected accrue to the city in the form of general purpose grants. Finally, there is a general purpose grant- in-aid under the Motor Vehicles Tax Act. Before 1939 a toll tax was levied on vehicles and animals entering the Municipal limits. This tax has been abolished, but a general purpose grant is still distributed based on the average of receipts of the three years preceding 1939. Both the grant under the motor vehicles act and the assessment and land revenue grants are small in amount and both have declined as a fraction of total grants (see Table 26) t Education assistance. The major form of state government assistance to the AMC is grants-in-aid for education purposes. These are of three kinds: primary education grants, secondary education grants, and grants for the Municipal medical college. The primary school grant is the largest received by the AMC and .accouts for nearly 60 percent of grants-in-aid. Under the two components of the grants, 50 percent of 'fadmissable" expenditures on compulsory education (grades I - 4) and 25 percent of "adissable" expenditures on non-compulsory education (grades 5-7) are reimbursed through the grant. The "admissable" expenditure cononents include: (a) Salaries up to-state government approved levels, including provident fund and other benefits (b) Salaries/of Administrat-ive. Officers (c) Building rents 68. Table 25: RATE OF TEEEUCATION CESS BY RATEABLE VALUE CLASS Rateable Vul.ue Crouo Residential Rate Non-Residential Rate (in Rupees) 0 - 300 Exempt Exempt 301 - 1000 3 percent L5 percent 1001 - 2000 5 percent 7.6 percent 2501 - 4500 6 percent 9.0 percent over 4500 7 percent 10.5 percent -Table 26t GRANTS FROM THE STATE GOVERNMENT TO THE AIMEDABAD LOCAL GOVERNMENTS 1965-66 1966-67 1967-68 1968-69 1969-70 1970-71 1971-72 Total Grant in Aid (rupeeu) 11,10,614 41,29,0611 h8,91,0h2 57,10,060 71,75,3h9 97,4o,1111 113,66,226 Total as a Percent of 8.7 7.0 8.4 6.7 8.1 11.1 11.h Total Current. Revenue The Percentage Distribution of Granits in Aid: Genmerl Purpon 20.hl 20.32 22.07 2h.17 20.18 27.63 19.32 .E.U:Ation eiGS 7.94 11.52 10.61 14.31 12.68 22.65 1li.79 Asxsssmnt & Land levenue 12.19 B.li9 11.20 9.62 7.17 li.85 h.43 Vehiic3e Act .28 .30 .25 .22 .17 .13 .11 *:iuceLion 67.56 61.91 69.01 63.29 61.74 57.2! - 59.19 >riju;r~ i ucatidn -61.80 60.20 67.45 61.31 62.71 55.05 57.63 GOirls Ilh School 1.23 1.71 1.-6' 1.83 . 2.03 2.19 1.56 SIT il1!L lhintcipal ledical College h.53 --- /15 .13 -- - -- Heit-h 8.23 13.55 h.55 ·5.60 5.39 8.59 13.58 a rPnring --- 2.1! -- .70 .6 2.61 7.59 Auxliary Iurse Care .27 --- --- --- --- --- .75 H:GP Scherie 50 Expenditura 7.96 11.10 4.55 4.89 4-.711 5.98 5.214 Courtsa nd Polee 2.47 2.72 2.79 2.33 -8.36 4.011 5.95 rin,? R1eliz.ed and Credited to Govt. Trieamy under NG and Oer Acts -1.67 2.05 -2.79 2.33 8.211 2.95 ~ -.88 Tr.fSic Signal --- .67 --- --- .12 • g.09 1.06 0 0U1rý r 1.33 1.50 1.141 5.31 1.110 2.19 1-117 es under Shops ESTT Act .58 .79 .69 .51 -- 1.02 C1u -07 11 .07 n Vidhyla1ya .1 5 25 .16 .f1 .07 .07 ro rs . -- .25 -92 -10- - 1c ovyance Act 639 .33 .28 .22 .07 .10 Public ~ ~ ~ Covync c 70. (d) approved repair and maintenance (e) approved furniture expenditures (f) approved book and state expenditures (up to Rs.5 per ,pupil in the first year and Rs.1 per pupil thereafter. (g) 'approved petty expenses The non-compulsory- grant includes, as admissable expenditures: (a) approved salaries (b) 15 percent of rental expenditures (c) 20 percent of all other expenditures The actual grant-in-aid is 25 percent '.f '(a)-(c), minus 97.5 percent of what in theory, could be collected in fees and tuition.1/ The 97.5 percent is then returned as a part of the grant. More specifically, the non-compulsory grant received (G) may be defined from G = .25(A - .975F) + .975F = .25A + 173F where A = admissable expenditures F = "theoretical" tuition charges The result of these distribution formulae is that only about 30-35 percent of the cost of primary educatiori in Ahmeda'ad is covered, as opposed to 55-65 percbnt for other administrative units within the state. This allocation is a result of the spending which is not classified as admissable , and which is more prevalent in Ahmedabad than in the other cities of the state--pre-primary schooling, the teaching of English in grades 5-7, teachers' salaries which exceed the State standard, and experimental programs which have the general approval of the Ministry, but not approval for state financial support. 1/ In reality, this provision is ignored and fees and tuition are not collected. 71. A state government grant is also made for secondary education, specifically for the operation of a girls secondary high school. Under this grant, 45 percent of approved expenditures on the operation of the school are reimbursed. Finally, during the third plan period a grant was made for the Muni6ipal Medical College with a recurring annual amount of Rs.2000 per pupil (up to 100 pupils) and a non-recurring grant per pupil of Rs.15000. This grant has been discontinued. Health Assistance. Three specific grants are made in the general area of health services, all of which are Central Government schemes. There is a centrally aided scheme for the prevention of malaria under the National Malaria Eradication Program. The GOI has sanctioned a grant-in- aid of 50 percent of approved (current and capital) expenditures. Finally, in the area of family planning,-the Central Government provides a 100 percent grant against approved expenditures in the maintenance of a family planning bureau. There is also an ad hoc grant of Rs.1500 per annum for each maternity home having a centrally sponsored family planning scheme. Police-Justice Assistance. The maintenance costs of magisterial courts is borne entirely by the AMC. Taxes which are levied and collected b.y these courts are a state government revenue lwut are turned over to the AMC as a grant (with a one year lag). If traffic signals are installed and erected in the city as per instructions of the Traffic Advisory Committee, 50 percent of the cost is borne by the State government _police department). 72. Other Grants. There are a number of grants, smaller in amount, which deal with specific programs. The AMC is given a grant to implement the Shops and Commercial Establishments Act, with the grant amount being the lesser of fines collected under this Act or expenditures incurred in implementation. If there is a deficit, it is borne by the AMC. Other grant programs include those for the Census, and for public conveyances. Equalizing Features of Intergovernmental Flows A possiblity for describing the area allocation of funds by the Gujarat State government is to analyze the relationship between the per capita grant accruing to each municipality and the per capita level of taxes raised from own sources. One argument for the use of grants-in-aid is that of equalization of service levels, or of income. Another has to do with the relationship between grants and local revenue effort. Personal income data for individual cities are not available. It is possible, however, to examine the degree to which grants-in-aid reduce the interurban disparity in per capita revenues raised from internal sources. Per capita expenditures of municipal governments in Gujarat state vary about a mean of 44.09 rupees per capita for 1968-69. Among the 15 revenue districts, Ahmedabad ranks thirteenth, with a per capita expenditure of 30.81 in 1968-69. Per capita grants vary about a mean of Rs 6.03, with Ahmedabad's Rs 4.37 ranking tenth. If the distribution of grants among municipalities is intended to be expenditure-equalizing, there should be a negative and significant relationship between grants in aid and revenues raised from own sources. In order to determine whether grants-in-aid are equalizing, two relationships 73. are estimated here for the 52 municipalities in Gujarat State for fiscal year 1968-69 (see Table 27). The first relationship tests whether grants- 'in-aid per capita are related to tax effort, specified here as revenue from own sources divided by assessed value of property. V = a1 + b1 (Vr) (19) where Gv = per capita grants-in-aid 1968-69 Vr = revenue from own sources per rupee of assessed value of property in 1968-69. Ordinary least squares was used to estimate equation (19) with the following results: 2 G = 9.5- 0.75 (Vr) = .03 (20) (1.07) The relationship between per capita aid and tax effort is not significant, i.e. state grants are neutral with respect to tax effort. There apparently is neither incentive nor penalty for Local units to raise additional taxes1.- The second relationship tested is that between per capita grants- in-aid and taxable capacity, measured by assessed value per capita. Symbol.- ically, GP a2 + b2 (V) (21) where V = per capita assessed value in 1968-69 for each municipality p G = per capita grants-in-aid in 1968-69 for each municipality. p Ordinary least squares is used to estimate equation (21) with the following results: G = 7.4 + 0.005 (V ) - .04 (22) (1.5), 1/ It should also be noted that 13 of the observations had to be omitted from the analysis because their assessed value in thousands was zero, and divisiod'by zero is not possible (See Table 27). 74. Again, the regression coefficient is not significantly different from zero at the 0.1 level of significance. The municipalities with assessed value of less than one thousand rupees were omitted and equation three was re- estimated with the following results; G= 8.1 + 0.004 (V ) R2 = .04 (23) (1.1) p and again, the t-statistic is not significant at the 0.1 level of signifi- cance. If the parameter relating per capita grants and per capita tax- able capacity or assessed value was positive and significantly different from zero, one could say that per capita grants are not only non-equalizing but counter equalizing. Since the parameter is not significantly different from zero one may conclude that grants-in-aid among municipalities are probably not equalizing, and the re ults of the two relationships, equations (20) and (23), are consistent. In general it would seem that there is no effective guiding ratio- nale in the distribution of state government grants. 75. Table 27: REVENUE FROM OWN SOURCES, GRANTS IN AID, ASSESSED VALUE OF PROPERTY, AND POPULATION FOR 52 MUNICIPALITIES IN GUJA2AT STATE, 1968-69 Revenue from own Grants Assessed Popu- Sources in Aid Value lation ('000 Rs.) ('000 Rs.) ('000 Rs.) (in units) lHI ILK å, 2 n ' yRa AOs 1716A 170, AMkFLT 750 1303 970613 3464 PAL ANIPHR 1179 3156 2?777 2q13< DAKHO 1 8537 416A 20335 30P41 iHAVNAGAR k94 1Ap92 0 17109 8 AT A4 AbSAAS ?332 0 ?1A MAHUVå 18976 0 167? 3166A PALITANA 5643 A36A 1?373 24591 S AVA RKNDL~ A9777 419(6 13952 301?2 ANKLESHWARI 7875 248 19620 ?0297 IRn4CH 27033 8s36 734440 73 9 kAJP1PLA 6192 2457 13460 21426 KILTMRA 12?202 241R 13936 2491 I1HLSåR 16945 3215 24009 3505P NavS ARI 29373 9344 56573 51300 JAMNAGAR 63501 8100 I 07QA5 1369? JU>NA GMH 2T2 420Q 0 74299 M4NGRIOL 368? 799 4287 2109 pnR\ANr) åR 27840 1580 4A769 71476A VFRAVAL PRAS 23397 250 0 605A4 ANIAND 24335 3145 5 40458 NCIRS AD) 6172 1993 9207 24704 CAMBAY 15804 24?0 37427 51291 KAPAnWANIJ 8035 3946 12691 77053 NADTån 32224 1022 Anson .7R952 PETLAD 34335 198? 1701 36239 UMR ETH 34Al 1/35 11490 21 24 ANJAR 4597 ~T?'4 ~~~l- TW3 ' 30T- 9J56,51 1755 0 3-953 . i~ *A:m 4750 ii36~ 46497 26,446, MANDVI 3071 4Aq 9667 2660 KAD) 4624 1568 0 23A66A1 KALAL 14649 2163 30928 31940 MFHSANA 10875 0 204?6 32577 PATAN 11377 2014 29437 50?64 SInHPUR '~4732 7718 29457 33850 INJH A 7880 3073 15092 70371 VISNAGAR 13576 1389 15459 259R onHAD 9063 2926 2177? 35493 GcDHRå 12282 A?7R 30770 521A7 OHnR AJ I 13059 570 0 49951 G1NiAL 17632 2871 30147 150AQ JETPUR - 13075 7479 0 311R6 MORVI 11835 1142 3558 501Q? RAJKnT 87460 10370 177936 19348 IIPLFTA 1171A 4?29 0 27613 WANKANER 5350 1337 9295 20231 1)HRANiGADHR A 3497 375 0 3?197 L1MK01 5790 373 0 2180 SIRHNIDRANAGAR 17995 3779 0 4qAo? WA0AWAN 2875 115, 0 271n4 Source: Statistics of Municipal Towns and Cities, Gujarat 1968-69, Bureau of Economics and Statiptics, Governri.nt of Gujarat. - 76 - VII. Problems and Policy Recommendatibns This section briefly summarizes the major problems of local fiscal management in Ahmedabad and presents some recommendations which may help alleviate them. The purpose of the present report is not to present a blue print for local (iscal reform, and thus the policy con- clusions are offered only as'a guide to future studies of fiscal reform. Bearing in mind these caveats, there are three general areas where discretionary policy reforms might improve the system of metropolitan finances in Ahmedabad: budget structure and administration, expenditure planning, and tax structure reform. The objective of these reforms would be to improve the overall efficiency of local government operation in Urban Ahmedabad through better planning; to make the distribution of tax burdens more reflective of differences in ability-to pay; and to raise the level of resources available for the financing of public programs. Though workable discretionary actions to achieve these goals may be taken-by Ahmedabad local governments, it should be emphasized that there are also local government finance problems in metropolitan Ahmedabad which require discre- tionary actions by higher levels of government. We do not consider State/' Federal government remedial action'in this sedtion -- the implications - 77 - of such reform are far to sweeping to be dealt with here. Instead, we focus on those actions which local.governments might take directly and within a relatively short period of time. Budget Structure and Administration The budget document, as prepared by Ahnedabad local governments, is not structured so as to be a guide in intermediate and long-term fiscal planning. The budget contains no clear breakdown of expenditures by function, estimates are prepared only for-a one year period, and the preparation of revenue estimates and the preparation of the overall budget are the respons- ibility of two different divisions within the city administration. Moreover, there is no consolidated local government budget which shows the aggregate fiscal position of the Ahmedabad Municipal Corporation (AM4C), Ahmedabad Metropolitan Transit Corporation (AMTS) and the Education special distric. This budget structure and process is indicative-of the fragmented approach to fiscal planning and evaluation which has been taken in the past. There is no long term fiscal planning by Ahmedbad's local govern- ments. Revenues are not estimated for a period longer than one year, and there exists no machinery or procedures for revenue forecasting. There is no medium-term capital budget, and it appears that capital expenditure. decisions are made on a year-to-year basis. The absence of a capital budget means that capital facilities planning in metropolitan Ahmedabad is not a continous process, that tax reforms are ad hoc and not planned for, and in general that there is no translation of long term allocation and distribution goals into a fiscal planning document. Given these conditions, the following recommendations may be made. First, the AMCGeneral Board should consider taking steps to create a Capital Budget, to regularize its updading, and to formalize the use of -78- such a budget in the fiscal planning process. The capital budget should cover the activities of all local governments in the Ahmedabad area and extend approximately over a five-year period. The goals of this budget are a forecast of the balance between current revenues and expenditures, the formulation of a set of capital project priorities, and the detailing of a financing plan for capital expenditures. The capital budget can serve as the primary fiscal planning tool of local governments and will enable an assessment of the fiscal consequences of a variety of government actions:4 e.g. alternative collective bargaining agreements, changes in formal State- Local relations, etc. Second, the AMC General Baard should consider taking steps to create and fund a Fiscal Planning Division in the office of the Municipal Commissioner. The fiscal planner should be a high level technician, supported by clerical assistance and part-time.-staff'-- perhaps from the University. The duties of the Fiscal Planning Division (FPD) would include (a) the preparation and updating of a capital budget, (b) the preparation of an annual "fiscal problems" report to the General Board, and (c) the carrying out of special studies and assignments at the request of the General Board. Properly used, the Fiscal Planning Division can condense large amount of material for the Board and can markedly improve the fiscal planning and decision making process. Third, the General Board should consider commissioning a study of budget procedures which would address the issue of how best to restructure the budget document. The .document should clearly distinguish between current - 79 - and capital expenditures, and should be divided according to a meaningful set of expenditure functions (see'text above). The goal of such reform would be to create a useable planning document. Expenditure Planning There is need to reconsider the rationale behind the distribution of AMC expenditures. The Milk and AMTS enterprises run deficits and there is no fiscal, or distributional, plan'which dictates the choice between continued subsidy and increased user charges. A useful first step in deriving an expenditure plan would be for the Municipal Corporation, together with the AMTS, and in consultation with the Gujarat State government, to commission a study to recommend a formal statement of a long term strategy for the distribution of current government expenditures. Such a strateg7 would enable government fiscal planners to rationalize the financing of enterprise type activities, e.g., bus service, water supply, milk and refuse collection. The issue is whether these services should be financed through user charges and whether such user charges should cover marginal cost., or whether there should be general revenue .fund subsidies on the grounds that there are income distributional reasons for such subisidies. If the latter position is taken, then it must be demonstrated that the property tax/octroi incidence is more regressive than the pattern of user charges. The two areas where an increase in user charges would seem possible are transportation (bus fares) and the Milk distribution scheme (milk charges). However, it seems clear that a general revenue subsidy to the Milk Scheme could be justified on distributional grainds. It is less clear that subsidies to the AMTS haye favorable distributional effects, i.e., increased - 81 - The rate structure of the property tax should be re- examined in light of inflation and the relatively slow growth of AMC revenues. By 1973, over one-fourth of all residential rateable value was in the top two rateable value classes. Continued inflation, reassessments, and the removal of preferential assessments for owner-occupiers could push this percentage to well over half-in the next few years. If the progressivity and the rate elasticity features of the system are to be retained, the rate structure should be changed by differentiating and subdividing the rabove Rs 5000" rateable value class. * APPEDIC A TE GROWTH RATE OF REAL PER1 CAPITA INCOME IN GUJARAT STATE BETWEEN 1964 AND 1971 The purpose of this appendix is to illustrate and describe the use of the differential technique employed in the text. First select the initial year (1964) as the base year for prices. The price index for 1971 with 1964 as the base year is 1.76. Next, note that total income is equal to the population multiplied by per capita in- come multiplied by price in the base year. Total income in the final year, 1971 (Y) may be written as the following identity: T = (P + dr) (I = dl) ( pr +.cdYpr) (1) where P - population in the base year, 1964 - I = Price in the bas.e year, or 1.00 y = real per capita income in the base year, which is simply pr per capita income in the base year d = the change in population between the base and the last year and similarly for the other two variables. Since all of the initial year variables are known and the change in population and prices are also IEown between 1964 and 1971, we can solve for the only unknown, that is, the change in real per capita income (dI,) between 1964 and 1971. If the total income in the base year is subtracted from both sides of the equation one can then find the difference in total income between the initial and final years: Appendix A Page 2 dy (P + dP) (I +dl) (Ypr + dYpr) -(Pi #,pr) (2) where dy is the change in real per capita income between the base and final years, 1964 and 1971. Substituting the values of the variables listed in Appendix Table 2 into equation (1), we obtain: 20,77,02,66,000 = (26,697,000) (1.75) (41o + dypr) (2a) 445 = 4lo + dYpr 35 = dzpr The percentage increase in real pe capita income in Gujarat State between 1964 and 1971 in 1964 prices is 35/110 or 8.5 percent. The question which is addressed in the text is what percentage of the growth in total income is attributable solely to growth in Real income, population and prices considered separately. More specifically if population and prices remained constant and real income increased by 3 rupees, what percentage of the actual increase in total income would, this represent. In order to address this question consider equation (2) above. d = (P +dP) (I + dl) (Ypr r - (P-IYpr) (3) Expanding the right side of the above expression one obtains: dYI pr dF ' +P b pr + ,Ypr rdl + P - d2 Sdr + I.*dP'dY +YT .dl*dP dP'dI1'dY pr pr pr pr The first three terms of equation (4) represent the change in total income that is due solely to changes in population, real per capita incom, and prices respectively, which the-last four terms represent the interaction of all three effects with each other. Letting T represent the interaction terms (4) becomes: Appendix A Page 3 dY=I, rprt dP +P+ pr Y dIpr + T (5) Substituting the values in Tables 2 and 35 for A RPCI 11,66,09,95,160 = (1.00) (410) (4,479,266) + (22,217,734) (1.00) (35) + (22,217,734)(410)( 75) + T (6) multiplying the expressions on the right hand side of equation (6) one obtains: 11,66,09,95,160 = 1,83,64,99,060 + 77,76,20,690 + 6,84,30,62,072 + T (7) Thus, if population and prices remained at their 1964 levels and real per capita income grew by 35 r(.pees, then this would account for 77,76,20,690/ 11,66,09,95,160 or 6.7 percent of the actual increase. Similarly if prices and per capita income remained at their 1964. levels and only population grew between 1964 and 1971 the increase in total income would have been 1,83,64,99,00/1L,66,09,95,160 or 15.7 percent of the actual increase in total. income. Finally, if real per capita income and population remained at their 1964 levels and prices increased by 75 percent then total income would have increased by 58.7 percent of the actual increase. The interaction of all three effects accounted for the remaining 18.9 percent of the actual total increase in total income. ADendix A Table 1 A SUMMARY OF REVENUES AND EXPENDITURES IN ABM4EDABAD, 1961-72 ( in Rupees ) Revenue from own Grants Total Year Sources in Aid Revenue Expenditure Population 1961 395,31,930 .25,43,797 420,75,727 399,21,363 1,149,917 1962 451,13,711 28,41,398 479,55,109 476,91,600 1,187,864 1963 490,35,056 32,30,785 523,15,841 500,08,879 1,227,064 1964 547,78,823 48,63,469 596,42,292 572,17,552 1,267,557 1965 584,36,623 53,63,790 638,00,413 674,57,436 1,309,387 1966 :98,34,868 45,54,108 643,88,976 767,14,614 1,352,597 1967 683,78,062 41,34,108 725,12,170 797,92,103 1,397,233 1968 677,78,071 51,27,275 729,05,346 884,19,705 1,443,342 1969 958,25,929 57,15,796 1015,41,725 965,64,466 1,490,973 1970 1000,21,180 71,79,223 1072,00,403 975,70,309 1,540,174 1971 993,10,739 94,82,592 1087,93,331 1058,76,133 1,588,379 1972 1132,94,766 105,07,272 1238,02,038 1130,47,209 1,640,795 /a Average Annual 9.5 11.4 9.6 9.4 3.3 Percentage Rate of Growth /a Projected, using population growth between 1961 and 1971. ADpendix A Table 2 PER CAPITA INCOME IN GUJARAT STATE IN CONSTANT AND CURRENT PRICES FOR 1960-61, 1970-71 Per Capita Income Real Per Capita Income Year in Current Prices in 1960-61 Prices Prices Foulation /1 1961 362 362 1.00 20,631,375 1962 395 387 1.02 21,147,159 1963 388 374 1.04 21,675,838 1964 410 388 1.06 22,217;734 1965 488 411 1.19 22,773,177 1966 491 371 1.32 23,342,507 1967 554 370 1.50 23,926,069 1968 641 396 1.62 24,524,221 1969 592 360 1.64 25,137,327 :1970 671 379 1.77 25,765,760 1971 778 419 1.86 26,697,000 Average Annual Percentage Growth Rate 7.6 1.4 .6.2 .2.5 1/ Population was interpolated between 1961-1971 .using the compond growth rate of population between 1961-71. Source: Socio-Economic Review, Gujarat State 1972-73, Bureau of Economics and Statistics, Government of Gujarat, p. 40-41.